fix: switch GMW AI source from omniroute to 9router

- Change AI_LLM_BASE_URL default from omniroute.imrnes.team to 9router.asepharyana.my.id
- Update AI_LLM_MODEL default from 'text' to 'claude-opus-5' (bare model name
  compatible with 9router/OpenAI-compatible router)
- Update .env.example and inline comments to reflect 9router
- discord-gateway config now matches backend (which already uses 9router)

.env.example

@@ -84,9 +84,9 @@ BACKLOG_SYNC_BATCH_SIZE=100             # Messages per backlog batch, max 100 (d
 
 # === AI Analysis ===
 AI_ANALYSIS_ENABLED=false               # Enable AI content moderation (default: false)
-# AI_LLM_API_KEY=                        # REQUIRED if AI_ANALYSIS_ENABLED=true. LLM API key
+AI_LLM_API_KEY=                        # REQUIRED if AI_ANALYSIS_ENABLED=true. LLM API key
-AI_LLM_BASE_URL=http://100.121.180.82:20128/api/v1  # LLM API base URL (omniroute on imrnes; /api/v1 exposes OpenAI-compatible chat+embeddings)
+AI_LLM_BASE_URL=https://9router.asepharyana.my.id/v1  # LLM API base URL (9router — OpenAI-compatible router, replaces omniroute)
-AI_LLM_MODEL=text                       # LLM text model name (default: text)
+AI_LLM_MODEL=claude-opus-5                   # LLM text model name (default: claude-opus-5)
 # AI_LLM_VISION_MODEL=                   # Vision model for image analysis (falls back to AI_LLM_MODEL)
 # AI_LLM_EMBEDDING_MODEL=                # Embedding model for semantic moderation cache (optional; enables near-duplicate text reuse to save LLM calls)
 # AI_LLM_EMBEDDING_MIN_SIMILARITY=0.97   # Min cosine similarity to reuse a cached verdict (default: 0.97)

.hermes/plans/2026-08-18_gmw-explainability-semantic-search.md

@@ -0,0 +1,500 @@
+# GMW — Moderation Explainability (#1) + Semantic Search (#3) Implementation Plan
+
+> **For Hermes:** Use subagent-driven-development to implement task-by-task.
+> Hard constraint from user (2026-08-18): web is PUBLIC, read-only, for USERS not admins. Moderation MUST stay FULLY AUTOMATIC. Rules stay in CODE (no per-channel config UI).
+
+**Goal:** Make GMW transparent (users see why a message was moderated) and searchable (users can semantic-search the message corpus), via two fully-automatic, code-driven, read-only-public features.
+
+**Architecture:**
+- **#1 Explainability:** Persist the structured moderation verdict that already exists in `AnalysisResult` (`flags[]`, `categories[]`, `severity`, `confidence`, `evidence[]`) into new columns on `moderation_actions`, surface them through the existing public moderation oRPC + the existing public `moderation` dashboard view. No new behavior — only new *data* + new *read* paths.
+- **#3 Semantic Search:** Add a SECOND persistent Qdrant collection (`gmw_message_archive`) keyed by message id (NOT the TTL cache). Embed each captured text message at capture time (reuse `embedText`) and upsert. Add a public `messages.semanticSearch` oRPC + a read-only search UI on the public `messages` view. Best-effort / non-blocking — embed failures never affect moderation or capture.
+
+**Tech Stack:** TypeScript (discord-gateway + backend + frontend monorepo), Drizzle ORM + Postgres (PgBouncer on imrnes), Qdrant (100.121.180.82:6333), Next.js 16 App Router + shadcn/ui, oRPC over `/trpc`. pnpm. Deploy via GitHub Actions Nix build + `systemctl restart`.
+
+**Critical existing facts (verified in repo):**
+- `AnalysisResult` shape (`src/modules/ai-moderation/ai-analysis-worker.ts:55`): `messageId, status, flags[], categories[], severity, confidence, recommendedAction, score, analysis, correctedFlags?`. The shared `AnalysisResult` (`src/shared/moderation-types.ts:142`) ALSO has `evidence?: string[]` and `policyVersion?: string`. **THESE ARE ALREADY COMPUTED but only logged, never persisted to `moderation_actions`.**
+- `moderation_actions` schema is DEFINED TWICE with a divergence:
+  - `src/shared/database/schema.ts:638` → `pgModerationActionsTable` (authoritative, has `reset_nickname` in `action_type` enum).
+  - `src/shared/database/schema/messages.ts:25` → another `pgModerationActionsTable` (NO `reset_nickname`; gateway-local copy).
+  - The gateway's `ModerationActionsDb` (`src/modules/message-capture/moderationActionsDb.ts`) imports from `schema.ts` (the authoritative one). The `messages.ts` copy appears UNUSED for DB ops — but WE MUST ADD NEW COLUMNS TO BOTH to avoid type drift, OR confirm the `messages.ts` copy is dead and delete it. **Decision: add columns to `schema.ts` (authoritative) AND the `messages.ts` copy to keep `$inferInsert`/`$inferSelect` in sync (the gateway `ModerationAction` type flows from shared).** Line truncated
+- Migration mechanism: Drizzle-managed via `drizzle/migrations/*.sql` (journal `_journal.json`) applied by `runMigrations()` → `migratePostgres`. **New tables/columns must be added with `drizzle-kit generate` to produce a numbered `.sql` + journal entry**, OR (simpler, matches `0013_rename_*.sql` manual style) write a raw idempotent `.sql` under `drizzle/migrations/` AND register it in `_journal.json`. **Preferred here: use `pnpm drizzle-kit generate` so the journal stays consistent.** The leLine truncated
+- **Historical lesson (MUST respect):** a prior migration (`0004_drop_unused_ai_columns.sql` = old `001_drop`) DELETED `ai_evidence`, `ai_policy_version`, `ai_moderation_raw` from `messages` with the note "written but never read". → Our new `moderation_actions` columns MUST be read (serializer + FE render). No write-only columns.
+- Embedding client: `embedText(text)` / `embedTexts(texts[])` in `src/modules/ai-moderation/embeddingClient.ts`. Returns `null` if `AI_LLM_EMBEDDING_MODEL` not configured. Reuses `config.AI_LLM_BASE_URL` + `config.AI_LLM_API_KEY`. OpenAI SDK v6 → `encoding_format: "float"` REQUIRED (Nvidia rejects base64).
+- Qdrant client: `src/modules/ai-moderation/qdrantClient.ts`. Has `ensureQdrantCollection(vectorSize)`, `upsertQdrantPoint(cacheKey, vector, payload)`, `searchQdrant(vector, limit, scoreThreshold)`. These are hardcoded to the cache collection name `config.QDRANT_COLLECTION ?? "gmw_text_moderation"`. **#3 needs a second collection** → generalize the client to accept a collection name param (add `ensureQdrantCollectionV2(name, size)` / `upsertQdrantPointV2(name, id, vector, payload)` / `searchQLine truncated
+- Capture hook: `captureMessage()` (`src/modules/message-capture/messageCapture.ts:201`) calls `messageStore.upsertMessageForCapture(messageRecord)` then (if not backlog) `queueMessageAnalysis`. **#3 embed must happen here**, async + fire-and-forget, after successful insert.
+- Public moderation view: `services/frontend/src/app/(dashboard)/moderation/view.tsx` renders `ActionRow` per action. The `ModerationAction` FE type is in `services/frontend/src/lib/types/moderation.ts` (NO new fields yet). The backend `moderationService.listActions` SQL is in `services/backend/src/modules/moderation/moderation.repository.ts:68` (raw SQL, selects fixed columns, joins `messages`).
+- oRPC wiring: `services/backend/src/orpc/router.ts` → `moderationRouter` (stats, actions) and `messagesRouter` (list, byChannel, getById, review, attachments). New procedures added here.
+
+---
+
+## TASK 1 — Schema: add explainability columns to `moderation_actions`
+
+**Objective:** Persist structured verdict on moderation actions so it can be surfaced (read) later.
+
+**Files:**
+- Modify: `services/discord-gateway/src/shared/database/schema.ts` (authoritative `pgModerationActionsTable`, ~line 638)
+- Modify: `services/discord-gateway/src/shared/database/schema/messages.ts` (`pgModerationActionsTable` copy, ~line 25) to keep type in sync
+- Create: `services/discord-gateway/drizzle/migrations/0015_add_moderation_explainability.sql`
+- Update: `services/discord-gateway/drizzle/migrations/meta/_journal.json` (add new entry)
+
+**Step 1: Add columns to both schema definitions**
+Add after `executed_at` in BOTH `pgModerationActionsTable` definitions:
+```ts
+    // ── Explainability (structured verdict, surfaced read-only to public web) ──
+    flags: pgText("flags"),            // JSON array of string flags, e.g. ["sara_agama","vulgar"]
+    categories: pgText("categories"), // JSON array of category strings
+    severity: pgText("severity", {
+      enum: ["none", "low", "medium", "high", "critical"],
+    }),
+    confidence: pgReal("confidence"),  // 0..1
+    score: pgReal("score"),            // 0..1 raw model score
+    evidence: pgText("evidence"),      // JSON array of short quoted snippets
+    policy_version: pgText("policy_version"), // rules.ts policy version string
+```
+Note: `flags`/`categories`/`evidence` stored as JSON-stringified TEXT (consistent with how `messages.ai_moderation_flags`/`ai_categories` are stored as TEXT elsewhere — confirm storage format in `updateMessageAIAnalysis`). Keep nullable.
+
+**Step 2: Generate/author the migration SQL**
+`0015_add_moderation_explainability.sql` (idempotent):
+```sql
+-- Add structured explainability columns to moderation_actions (read-only surfaced to public web).
+ALTER TABLE IF EXISTS "moderation_actions"
+  ADD COLUMN IF NOT EXISTS "flags" text,
+  ADD COLUMN IF NOT EXISTS "categories" text,
+  ADD COLUMN IF NOT EXISTS "severity" text
+    CHECK ("severity" IS NULL OR "severity" IN ('none','low','medium','high','critical')),
+  ADD COLUMN IF NOT EXISTS "confidence" real,
+  ADD COLUMN IF NOT EXISTS "score" real,
+  ADD COLUMN IF NOT EXISTS "evidence" text,
+  ADD COLUMN IF NOT EXISTS "policy_version" text;
+```
+Register in `_journal.json`: append an entry with `idx: 15`, a new unique `tag` (hash), `version`, `when` = Date.now(), `tag` short, `breakpoints: false`. Use `pnpm drizzle-kit generate` if possible to get a correct tag; otherwise hand-edit the journal carefully (copy an existing entry's shape).
+
+**Step 3: Type-check gateway**
+Run: `cd services/discord-gateway && pnpm typecheck`
+Expected: PASS (no new compile errors).
+
+**Step 4: Commit**
+```bash
+git add services/discord-gateway/src/shared/database/schema.ts \
+        services/discord-gateway/src/shared/database/schema/messages.ts \
+        services/discord-gateway/drizzle/migrations/0015_add_moderation_explainability.sql \
+        services/discord-gateway/drizzle/migrations/meta/_journal.json
+git commit -m "feat(db): add explainability columns to moderation_actions"
+```
+
+---
+
+## TASK 2 — Persist verdict at the auto-delete + command-handler call sites
+
+**Objective:** Populate the new columns from the already-computed `AnalysisResult` when a moderation action is logged. Fully automatic, no new behavior.
+
+**Files:**
+- Modify: `services/discord-gateway/src/modules/ai-moderation/autoDeleteManager.ts` (`logAutoDeleteAttempt` ~line 166, and the second `createModerationAction` call ~line 234 for nickname/mute paths)
+- Modify: `services/discord-gateway/src/modules/command-handler/moderation.handler.ts` (`createModerationAction` ~line 95)
+- Helper (create): `services/discord-gateway/src/modules/ai-moderation/verdictToActionFields.ts` — shared mapper so all 3 call sites stay DRY.
+
+**Step 1: Create the mapper helper**
+`verdictToActionFields.ts`:
+```ts
+import type { AnalysisResult } from "@/modules/ai-moderation/ai-analysis-worker";
+import type { ModerationActionInsert } from "@/shared/index"; // or inline shape
+
+/**
+ * Map a computed AI verdict into the explainability columns of a moderation
+ * action. Null-safe: missing fields stay null (e.g. manual admin actions have
+ * no AnalysisResult). This is read-only structured data — it does NOT change
+ * any enforcement decision.
+ */
+export function verdictToActionFields(result?: {
+  flags?: string[];
+  categories?: string[];
+  severity?: string;
+  confidence?: number;
+  score?: number;
+  evidence?: string[];
+  policyVersion?: string;
+}): {
+  flags: string | null;
+  categories: string | null;
+  severity: string | null;
+  confidence: number | null;
+  score: number | null;
+  evidence: string | null;
+  policy_version: string | null;
+} {
+  if (!result) {
+    return { flags: null, categories: null, severity: null, confidence: null,
+             score: null, evidence: null, policy_version: null };
+  }
+  const j = (v: unknown) => (v == null ? null : JSON.stringify(v));
+  return {
+    flags: j(result.flags),
+    categories: j(result.categories),
+    severity: result.severity ?? null,
+    confidence: result.confidence ?? null,
+    score: result.score ?? null,
+    evidence: j(result.evidence),
+    policy_version: result.policyVersion ?? null,
+  };
+}
+```
+
+**Step 2: Wire `logAutoDeleteAttempt`**
+Find the `createModerationAction({...})` in `logAutoDeleteAttempt` and spread the verdict fields:
+```ts
+await messageStore.createModerationAction({
+  message_id: message.id,
+  user_id: message.user_id,
+  guild_id: message.guild_id,
+  action_type: "delete_message",
+  reason: result.reason,
+  ...verdictToActionFields(result.analysisResult), // <-- pass the AnalysisResult through
+  executed_by: "auto-delete-manager",
+  status: ...,
+});
+```
+**IMPORTANT:** `result` here is `AutoDeleteResult` — verify it carries the `AnalysisResult` (or the verdict). If `AutoDeleteResult` does NOT carry the full `AnalysisResult`, trace where `attemptAutoDeleteFlaggedMessage` is called from and pass the `AnalysisResult` down (it is available in the analysis worker that triggered the delete). Confirm by reading `AutoDeleteResult` type + its producer. If the verdict is only available at the orchestrator level, add an optional `verdict?: AnalysisResulLine truncated
+
+**Step 3: Wire the second call site in `autoDeleteManager.ts`** (the mute/nickname path ~line 234) similarly, if it has an `AnalysisResult` available; otherwise leave fields null (manual-style action).
+
+**Step 4: Wire `moderation.handler.ts`** command path (~line 95) — pass `verdictToActionFields(verdict)` if the command handler has the `AnalysisResult` for the target message; otherwise nulls. Confirm what the handler receives.
+
+**Step 5: Type-check + lint**
+Run: `cd services/discord-gateway && pnpm typecheck && pnpm lint`
+Expected: PASS.
+
+**Step 6: Commit**
+```bash
+git add services/discord-gateway/src/modules/ai-moderation/verdictToActionFields.ts \
+        services/discord-gateway/src/modules/ai-moderation/autoDeleteManager.ts \
+        services/discord-gateway/src/modules/command-handler/moderation.handler.ts
+git commit -m "feat(mods): persist structured verdict into moderation_actions"
+```
+
+---
+
+## TASK 3 — Backend: surface explainability in `moderation.actions`
+
+**Objective:** Read the new columns in the public oRPC so the frontend can render them. (Read path — satisfies the "never write-only" rule.)
+
+**Files:**
+- Modify: `services/backend/src/modules/moderation/moderation.repository.ts` (`listActions` raw SQL ~line 68) — add new columns to SELECT + map.
+- Modify: `services/frontend/src/lib/types/moderation.ts` (`ModerationAction` interface) — add new fields.
+- Modify: `services/frontend/src/app/(dashboard)/moderation/view.tsx` (`ActionRow`) — render flags/categories badges + severity + confidence + evidence snippet.
+
+**Step 1: Extend backend SELECT**
+In `listActions`, add to the SELECT list: `a.flags, a.categories, a.severity, a.confidence, a.score, a.evidence, a.policy_version`. In the `.map(...)` add:
+```ts
+flags: r.flags ? safeJsonArray(String(r.flags)) : null,
+categories: r.categories ? safeJsonArray(String(r.categories)) : null,
+severity: r.severity ? String(r.severity) : null,
+confidence: r.confidence != null ? Number(r.confidence) : null,
+score: r.score != null ? Number(r.score) : null,
+evidence: r.evidence ? safeJsonArray(String(r.evidence)) : null,
+policy_version: r.policy_version ? String(r.policy_version) : null,
+```
+where `safeJsonArray(s)` = `JSON.parse(s)` wrapped in try/catch returning `[]` on failure (define a tiny local helper in the repository file).
+
+**Step 2: Extend FE type**
+In `services/frontend/src/lib/types/moderation.ts` `ModerationAction`:
+```ts
+  flags: string[] | null;
+  categories: string[] | null;
+  severity: "none" | "low" | "medium" | "high" | "critical" | null;
+  confidence: number | null;
+  score: number | null;
+  evidence: string[] | null;
+  policy_version: string | null;
+```
+
+**Step 3: Render in `ActionRow`**
+After the existing reason block, add (using existing `Badge` + `aiTone` from `@/lib/ai-status`):
+```tsx
+{a.severity && (
+  <Badge tone={aiTone(a.severity === "none" ? "clean" : a.severity)}>
+    {a.severity}
+  </Badge>
+)}
+{a.flags?.length ? (
+  <div className="mt-1 flex flex-wrap gap-1">
+    {a.flags.map((f) => <Badge key={f} tone="amber">{f}</Badge>)}
+  </div>
+) : null}
+{a.evidence?.length ? (
+  <div className="mt-1 text-xs text-ink-faint border-l-2 border-hairline pl-2">
+    “{a.evidence[0]}”
+  </div>
+) : null}
+{a.confidence != null && (
+  <div className="mono mt-0.5 text-[0.6rem] text-ink-faint">
+    conf {(a.confidence * 100).toFixed(0)}%
+  </div>
+)}
+```
+Keep `ActionRow` read-only. No admin controls.
+
+**Step 4: Type-check both services**
+Run: `cd services/backend && pnpm typecheck && pnpm lint` and `cd services/frontend && pnpm typecheck && pnpm lint`
+Expected: PASS.
+
+**Step 5: Commit**
+```bash
+git add services/backend/src/modules/moderation/moderation.repository.ts \
+        services/frontend/src/lib/types/moderation.ts \
+        services/frontend/src/app/\(dashboard\)/moderation/view.tsx
+git commit -m "feat(web): surface moderation explainability (flags/severity/evidence)"
+```
+
+---
+
+## TASK 4 — Qdrant client: support a second persistent collection
+
+**Objective:** Generalize the Qdrant client so #3 can use a dedicated archive collection without disturbing the automod cache.
+
+**Files:**
+- Modify: `services/discord-gateway/src/modules/ai-moderation/qdrantClient.ts`
+
+**Step 1: Add collection-aware variants**
+Refactor `collectionName()` to accept an optional name, and add V2 functions that take an explicit collection:
+```ts
+function collectionName(fallback = config.QDRANT_COLLECTION ?? "gmw_text_moderation"): string {
+  return fallback;
+}
+export const ARCHIVE_COLLECTION = config.QDRANT_ARCHIVE_COLLECTION ?? "gmw_message_archive";
+
+export async function ensureQdrantCollectionV2(name: string, vectorSize: number): Promise<boolean> {
+  // same body as ensureQdrantCollection but uses `name` instead of collectionName()
+}
+export async function upsertQdrantPointV2(
+  name: string, pointId: number, vector: number[], payload: QdrantVerdictPayload,
+): Promise<boolean> { /* PUT /collections/{name}/points with wait:true */ }
+export async function searchQdrantV2(
+  name: string, vector: number[], limit: number, scoreThreshold: number,
+): Promise<QdrantSearchHit[]> { /* POST /collections/{name}/points/search */ }
+```
+Keep all existing `ensureQdrantCollection` / `upsertQdrantPoint` / `searchQdrant` UNCHANGED (cache path). V2 functions mirror them with the `name` param. Reuse `request()` and the existing payload/score types.
+
+**Step 2: Type-check**
+Run: `cd services/discord-gateway && pnpm typecheck`
+Expected: PASS.
+
+**Step 3: Commit**
+```bash
+git add services/discord-gateway/src/modules/ai-moderation/qdrantClient.ts
+git commit -m "feat(qdrant): add collection-aware V2 upsert/search for archive"
+```
+
+---
+
+## TASK 5 — Capture-time embed + archive upsert
+
+**Objective:** Make every (non-backlog, text) captured message searchable in the persistent archive. Non-blocking / best-effort.
+
+**Files:**
+- Modify: `services/discord-gateway/src/modules/message-capture/messageCapture.ts` (`captureMessage` ~line 201)
+- Create: `services/discord-gateway/src/modules/message-capture/archiveEmbedder.ts` — wraps embed + upsert with fire-and-forget + rate-limit guard.
+
+**Step 1: Create `archiveEmbedder.ts`**
+```ts
+import { createChildLogger } from "@/shared/logger/index";
+import { embedText } from "@/modules/ai-moderation/embeddingClient";
+import { ARCHIVE_COLLECTION, ensureQdrantCollectionV2, upsertQdrantPointV2, qdrantPointId } from "@/modules/ai-moderation/qdrantClient";
+import { config } from "@/shared/config/config";
+
+const log = createChildLogger("archive-embedder");
+
+/**
+ * Fire-and-forget: embed a captured message and upsert into the persistent
+ * archive collection. Failures are swallowed — searching is a nice-to-have,
+ * never a precondition for capture or moderation.
+ */
+export function archiveMessageEmbedded(message: {
+  id: string; content: string; username: string; channel_id: string; guild_id: string; created_at: number;
+}): void {
+  if (!config.AI_LLM_EMBEDDING_MODEL) return; // embeddings disabled → skip
+  if (!message.content || message.content.trim().length < 3) return;
+  void (async () => {
+    try {
+      const vector = await embedText(message.content);
+      if (!vector) return;
+      const ok = await ensureQdrantCollectionV2(ARCHIVE_COLLECTION, vector.length);
+      if (!ok) return;
+      await upsertQdrantPointV2(ARCHIVE_COLLECTION, qdrantPointId(`archive:${message.id}`), vector, {
+        text: message.content.slice(0, 4000),
+        flags: "", // not a verdict payload; keep shape compatible
+        analyzed_at: Date.now(),
+        expires_at: Date.now() + 1000 * 60 * 60 * 24 * 365 * 5, // 5y persistent
+        content_hash: undefined,
+      });
+    } catch (err) {
+      log.debug({ messageId: message.id, error: err instanceof Error ? err.message : String(err) }, "archive embed skipped");
+    }
+  })();
+}
+```
+Payload type reuse: `QdrantVerdictPayload` has `text`, `flags`, `analyzed_at`, `expires_at`, `content_hash?`. For the archive we only need `text` + timestamps; set `flags: ""` (empty, ignored by search filter which keys on `expires_at`). **Acceptable:** the search path filters `expires_at >= now` — 5y window satisfies that.
+
+**Step 2: Call from `captureMessage`**
+In `captureMessage`, after `const inserted = await messageStore.upsertMessageForCapture(messageRecord); if (!inserted) return;` and BEFORE the backlog branch, add:
+```ts
+if (!isBacklog && messageRecord.content) {
+  archiveMessageEmbedded(messageRecord);
+}
+```
+(`messageRecord` is the `MessageRecord` from `buildMessageRecord`; confirm it carries `content`, `channel_id`, `guild_id`, `created_at`. It does — see `messagesCrud`/`types`.)
+
+**Step 3: Type-check + lint**
+Run: `cd services/discord-gateway && pnpm typecheck && pnpm lint`
+Expected: PASS.
+
+**Step 4: Commit**
+```bash
+git add services/discord-gateway/src/modules/message-capture/archiveEmbedder.ts \
+        services/discord-gateway/src/modules/message-capture/messageCapture.ts
+git commit -m "feat(archive): embed captured messages into persistent Qdrant archive"
+```
+
+---
+
+## TASK 6 — Backend: `messages.semanticSearch` oRPC
+
+**Objective:** Public, read-only semantic search over the message archive.
+
+**Files:**
+- Modify: `services/backend/src/modules/messages/messages.repository.ts` — add `semanticSearch(query, limit, guildId?)`.
+- Modify: `services/backend/src/modules/messages/messages.service.ts` — expose `semanticSearch`.
+- Modify: `services/backend/src/orpc/router.ts` — add `messages.semanticSearch` procedure.
+- Create (or reuse): an embedding call from the backend. The backend does NOT import the gateway's `embeddingClient`. **Decision:** add a minimal backend embed helper `services/backend/src/modules/messages/embed.ts` that calls the same OpenAI-compatible endpoint via `config` (reuse `config.AI_LLM_BASE_URL`/`AI_LLM_API_KEY`/`AI_LLM_EMBEDDING_MODEL` if present on the backend; if not configured, return a clear "search unavailable" error). Mirror `encoding_format: "float"`.
+- Modify: `services/backend/src/modules/messages/messages.schema.ts` — add `semanticSearchQuery` zod schema (limit, guildId?, query).
+
+**Step 1: Backend embed helper** (`embed.ts`)
+```ts
+import OpenAI from "openai";
+import { config } from "@/shared/config/index";
+import { createChildLogger } from "@/shared/logger/index";
+const log = createChildLogger("messages-embed");
+let client: OpenAI | null = null;
+function getClient() {
+  if (!config.AI_LLM_API_KEY || !config.AI_LLM_EMBEDDING_MODEL) return null;
+  if (!client) client = new OpenAI({ apiKey: config.AI_LLM_API_KEY, baseURL: config.AI_LLM_BASE_URL, maxRetries: 0, timeout: 30_000 });
+  return client;
+}
+export async function embedQuery(text: string): Promise<number[] | null> {
+  const c = getClient(); if (!c) return null;
+  try {
+    const r = await c.embeddings.create({ model: config.AI_LLM_EMBEDDING_MODEL as string, input: text, encoding_format: "float" });
+    return r.data[0].embedding;
+  } catch (e) { log.warn({ error: e instanceof Error ? e.message : String(e) }, "query embed failed"); return null; }
+}
+```
+
+**Step 2: Repository `semanticSearch`**
+```ts
+async semanticSearch(queryVector: number[], limit: number, guildId?: string) {
+  // Search archive collection, then join messages for text + channel.
+  const hits = await searchQdrantV2(ARCHIVE_COLLECTION, queryVector, limit, 0.6);
+  const ids = hits.map(h => h.cacheKey.replace("qdrant:", "")); // point id → we stored archive:<messageId>
+  // decode: qdrantPointId is a uint64; we need the original message id.
+  // SIMPLER: store message_id inside the payload too. → update archiveEmbedder payload to include `message_id`.
+  ...
+}
+```
+**REFINEMENT (important):** `qdrantPointId` is a hash, not reversible. So the archive payload MUST carry `message_id` (and `channel_id`, `guild_id`, `username`, `created_at`) so the backend can return full results without a reverse lookup. **Update `archiveEmbedder.ts` payload** to include those fields, and relax `QdrantVerdictPayload` (or create `QdrantArchivePayload`) to allow them. Then `semanticSearch` returns the payloads directly (already contain text + metadata) — no DB join needed, anLine truncated
+
+**Step 3: Service + router**
+`messages.service.ts`: `async semanticSearch(query: string, limit: number, guildId?: string)` → embed → `repository.semanticSearch`.
+`orpc/router.ts` under `messagesRouter`:
+```ts
+semanticSearch: os
+  .input(z.object({ query: z.string().min(1), limit: z.coerce.number().int().positive().max(50).default(10), guildId: z.string().optional() }))
+  .handler(async ({ input }) => {
+    const results = await messagesService.semanticSearch(input.query, input.limit, input.guildId);
+    return { results, nextCursor: null };
+  }),
+```
+
+**Step 4: Type-check + lint (backend)**
+Run: `cd services/backend && pnpm typecheck && pnpm lint`
+Expected: PASS.
+
+**Step 5: Commit**
+```bash
+git add services/backend/src/modules/messages/embed.ts \
+        services/backend/src/modules/messages/messages.repository.ts \
+        services/backend/src/modules/messages/messages.service.ts \
+        services/backend/src/modules/messages/messages.schema.ts \
+        services/backend/src/orpc/router.ts
+git commit -m "feat(api): public semantic message search over archive"
+```
+
+---
+
+## TASK 7 — Frontend: semantic search UI on `messages` view
+
+**Objective:** Public, read-only search box + results on the existing messages dashboard.
+
+**Files:**
+- Modify: `services/frontend/src/app/(dashboard)/messages/page.tsx` + `view.tsx` — add a search input (debounced) that calls a new `useMessagesSemanticSearch` hook → `messages.semanticSearch` oRPC, renders results as message cards (reuse `GlassPanel`/`Badge`/existing message row components).
+- Modify: `services/frontend/src/lib/types/message.ts` — add `SemanticSearchResult` + `SemanticSearchResponse` types.
+- Modify: `services/frontend/src/lib/api/client.ts` (or `server.ts`) — add `semanticSearch` fetcher/routers export if using oRPC client; if the FE uses raw fetch through the proxy, add a `POST /api/messages/semantic-search` or an oRPC client call consistent with existing `messages.*` calls (follow the EXISTING pattern in `src/lib/api/` — inspect how `messages.list` is called and replicate).
+
+**Step 1: Add FE types**
+```ts
+export interface SemanticSearchResult {
+  message_id: string;
+  content: string;
+  username: string;
+  channel_id: string;
+  guild_id: string;
+  created_at: number;
+  score: number;
+}
+export interface SemanticSearchResponse { results: SemanticSearchResult[]; nextCursor: string | null; }
+```
+
+**Step 2: Add hook + wire view**
+Follow the existing `use-moderation.ts` SWR pattern. Add `useMessagesSemanticSearch(query, guildId?)` returning `{ data, isLoading, error }`. In `messages/view.tsx`, add a search `Input` (from `@/components/primitives`) at the top, debounce ~300ms, and render results below the live list when a query is present. Reuse the message-row rendering already in that view (do not invent a new component).
+
+**Step 3: Type-check + lint (frontend)**
+Run: `cd services/frontend && pnpm typecheck && pnpm lint`
+Expected: PASS.
+
+**Step 4: Commit**
+```bash
+git add services/frontend/src/app/\(dashboard\)/messages/ \
+        services/frontend/src/lib/types/message.ts \
+        services/frontend/src/lib/api/ \
+        services/frontend/src/hooks/
+git commit -m "feat(web): public semantic message search UI"
+```
+
+---
+
+## TASK 8 — Build all services + deploy + verify
+
+1. `cd services/discord-gateway && pnpm typecheck && pnpm build && pnpm lint`
+2. `cd services/backend && pnpm typecheck && pnpm build && pnpm lint`
+3. `cd services/frontend && pnpm typecheck && pnpm build && pnpm lint`
+4. Commit any formatting fixes (biome `--unsafe` if import order), author `asepharyana`, NO Co-Authored-By.
+5. `git push origin main` → watch `gh run watch` on "Build & Deploy (Nix)".
+6. After deploy: verify `systemctl show gmw-discord-gateway.service --property=ActiveEnterTimestamp,SubState` reflects new timestamp; same for backend + frontend.
+7. Smoke: `curl -s http://127.0.0.1:4001/trpc/messages.semanticSearch?input=<urlencoded json>` OR via the public web `imphnen.asepharyana.my.id` messages page → type a query → expect results (after some messages have been embedded; embeddings only run on NEW captures post-deploy, so seed a few test messages or backfill).
+8. Moderation explainability: trigger/observe a flagged message → confirm `moderation_actions.flags` is populated (SQL `SELECT flags, severity FROM moderation_actions ORDER BY created_at DESC LIMIT 5;`) and the public moderation view shows badges.
+
+---
+
+## RISKS / TRADEOFFS / OPEN QUESTIONS
+- **Embedding cost:** #3 embeds EVERY captured message → more embedding API calls. Mitigated: only text ≥3 chars, fire-and-forget, skip if model unconfigured. If cost is a concern, batch embed (reuse `embedTexts`) per capture burst — but start simple (per-message) and observe.
+- **Backfill:** post-deploy, the archive is empty until new messages arrive. Optional follow-up: a one-off backfill script over existing `messages` (out of scope for this plan unless user asks).
+- **Schema duplication:** the `pgModerationActionsTable` double-definition must stay in sync (Task 1 patches both). If `messages.ts` copy is provably dead, a follow-up can delete it — but NOT in this plan (avoid scope creep / risk).
+- **`AutoDeleteResult` verdict availability (Task 2):** requires confirming the `AnalysisResult` is reachable at the `createModerationAction` call sites. If not, we add an optional field to `AutoDeleteResult` at the orchestrator call site. This is the highest-risk integration point — verify before assuming.
+- **Public exposure:** semantic search returns message text + usernames. This is INTENDED (web is public for users). No auth added. If a guild wants private, that is a future config (out of scope).
+- **Qdrant payload type:** reusing `QdrantVerdictPayload` for archive is slightly awkward (carries `flags`/`expires_at` semantics). Cleaner: introduce `QdrantArchivePayload` with `message_id`, `channel_id`, `guild_id`, `username`, `content`, `created_at`, `expires_at`. **Prefer the dedicated payload type in Task 4/5** to avoid confusion.
+
+## VERIFICATION CHECKLIST
+- [ ] `moderation_actions` has 7 new columns (DB + both schema defs).
+- [ ] A real auto-delete populates `flags`/`severity`/`evidence` (verified via SQL).
+- [ ] Public moderation view renders badges + evidence (manual browser check on imphnen.asepharyana.my.id/moderation).
+- [ ] New message capture upserts a point into `gmw_message_archive` (verify via Qdrant `/collections/gmw_message_archive/points/count`).
+- [ ] `messages.semanticSearch` returns relevant results for a known phrase.
+- [ ] All three services: typecheck + build + lint green; CI "Build & Deploy (Nix)" green; systemd timestamps updated.

.hermes/plans/2026-08-18_gmw-public-features-2-6.md

@@ -0,0 +1,101 @@
+# GMW — Fitur Publik Lanjutan (#2–#6) Implementation Plan
+
+> **For Hermes:** Implement task-by-task. Build + lint + typecheck each service
+> after its changes. Deploy via push to main (CI handles Nix build + systemd).
+> Hard constraint (user 2026-08-18): public read-only web, fully automatic,
+> rules in code, NO admin endpoints, NO shadow mode, NO per-channel web config.
+> **EXPLICITLY EXCLUDED: User Reputation / Strike History** (user: "hapus
+> sepenuhnya fitur user reputation" — it was never built; do not add it).
+
+## Existing infra to reuse (verified)
+- **WS**: backend `ws/server.ts` broadcasts JSON `{type,data,timestamp}` to
+  frontendClients. Backend `ws/redis-bridge.ts` subscribes Redis channels
+  listed in `DISCORD_CHANNEL_TO_WS_EVENT` (backend `shared/redis-channels.ts`)
+  and re-emits as WS events. FE `src/lib/ws` auto-reconnect typed client.
+- **Gateway → Redis**: `EventBroadcaster` + `RedisEventPublisher` (
+  `discord-gateway/src/modules/event-broadcaster`). Publish via
+  `eventBroadcaster.publish(EventChannels.X, payload)`.
+- **Moderation data**: `moderation_actions` table (now has explainability
+  cols). `moderation.repository.listActions` returns rows. `ModerationAction`
+  FE type at `frontend/src/lib/types/moderation.ts`.
+- **Messages**: `messages.list` / `getMessagesByChannel` (backend oRPC +
+  repository). FE `messagesApi` + `useMessages`.
+- **Charts**: NO chart lib installed. Use **pure SVG/CSS** (consistent with
+  repo; avoid new deps).
+- **CSV**: client-side Blob download, no backend.
+
+## Task 1 — Live Moderation Feed (#2)
+**Gateway**: add `MODERATION_ACTION: "discord:moderation:action"` to
+`redis-channels.ts` (shared) + `EventChannels.MODERATION_ACTION` in
+`eventTypes.ts`. In `moderationActionsDb.createModerationAction`, after insert,
+publish `eventBroadcaster.publish(EventChannels.MODERATION_ACTION, actionRow)`.
+**Backend**: add `DISCORD_MODERATION_ACTION` constant + map
+`[DISCORD_MODERATION_ACTION]: "moderation_action"` in `DISCORD_CHANNEL_TO_WS_EVENT`.
+**FE**: in `src/lib/ws`, subscribe to `moderation_action`; add `useLiveModeration`
+hook (SWR-style with WS push, capped buffer ~50). Add `<LiveModerationFeed>`
+client component on `/moderation` page (top of list, animated new-row).
+Risk: gateway publish at every action (already async insert) — fire-and-forget,
+wrap in try/catch. Verify WS event reaches FE via `wscat`/curl or log.
+
+## Task 2 — Toxic Topic Trends (#3)
+**Backend**: add `moderation.trends` oRPC. Query `moderation_actions` grouped
+by `categories` (jsonb text[]) over last 30 days, count per category + severity
+breakdown. Also `action_type` distribution. Return
+`{ categories: {name,count}[], severities: {level,count}[], actions: {type,count}[] }`.
+Map jsonb array in SQL (use `unnest` or parse in JS). Reuse `getDatabase`.
+**FE**: `useModerationTrends` hook + `<TopicTrends>` SVG bar chart (top 10
+categories) + severity donut (SVG arcs). Place on `/moderation` as a panel.
+
+## Task 3 — Channel Timeline / Replay (#4)
+Reuse existing `messages.list` (guildId) + `getMessagesByChannel`. Add a
+**Timeline tab** to `/messages` that groups messages by date (client-side
+bucket from `created_at`). Load-more via cursor. No new backend (existing
+`messagesRouter.list` already supports guildId+limit+cursor). If needed, add
+`messages.timeline` aggregation (count per day) — but keep simple: client
+groups fetched rows. Verify existing endpoint returns enough history.
+
+## Task 4 — Export CSV (#5)
+**FE only**. `lib/csv.ts` `toCsv(rows, columns)` + `downloadCsv(filename, csv)`.
+Add "Export CSV" button on `/moderation` (exports current actions) and
+`/messages` (exports current list). Pure client-side, read-only. No backend.
+
+## Task 5 — Activity Heatmap (#6)
+**Backend**: add `messages.activity` oRPC: per-channel message count grouped by
+hour-of-day (0–23) over last 14 days. Return
+`{ channels: {channelId, name, byHour: number[24]}[], max }`. Use SQL
+`EXTRACT(hour from ...)` + group by channel. Channel name from
+`message.metadata->'channel'->>'channelName'`.
+**FE**: `useMessageActivity` hook + `<ActivityHeatmap>` SVG grid (channels ×
+24h, color intensity = count/max). Place on `/messages` or `/dashboard`.
+
+## Verification checklist
+- [ ] `pnpm typecheck && pnpm lint && pnpm build` green for gateway, backend, frontend
+- [ ] Backend `/trpc/moderation/trends` returns categories/severities/actions
+- [ ] Backend `/trpc/messages/activity` returns byHour grids
+- [ ] WS `moderation_action` received by FE (log or visible live row)
+- [ ] No admin/write endpoint added; all public read-only
+- [ ] No User Reputation code anywhere (grep "reputation|strike|reputasi")
+- [ ] Deploy via push; all 3 services `running`; moderation + messages pages load
+
+## Files touched (summary)
+- gateway: `shared/redis-channels.ts`, `event-broadcaster/eventTypes.ts`,
+  `event-broadcaster/eventBroadcaster.ts`, `message-capture/moderationActionsDb.ts`
+- backend: `shared/redis-channels.ts`, `orpc/router.ts`,
+  `modules/moderation/moderation.service.ts` (+repository),
+  `modules/messages/messages.service.ts` (+repository, +schema)
+- frontend: `lib/ws/*`, `hooks/use-moderation.ts`, `hooks/use-messages.ts`,
+  `lib/csv.ts`, `lib/types/*`, `app/(dashboard)/moderation/view.tsx`,
+  `app/(dashboard)/messages/view.tsx`, new components under `components/`
+
+## Status: COMPLETE (deployed + verified)
+- Commit 9b3134d: features #2–#6 (live feed, trends, timeline, CSV export, heatmap)
+- Commit 2a8f6d9: user reputation feature fully removed (643 deletions, no trace in src/tests)
+- Migration 0016 applied: user_reputations DROPPED (DB verified: false)
+- All 3 services active (gateway + backend restarted 18:29, frontend running)
+- Gateway typecheck/lint/test(117 passed); backend typecheck/lint/build; FE lint/build — all GREEN
+
+## Verification
+- moderation/stats WS returns data (32 actions) → WS adapter works
+- DB: user_reputations gone; moderation_actions explainability cols present
+- Live Feed: gateway publishes discord:moderation:action → backend WS (same path as guild_member_*)
+- Trends/Activity: backend router procedures registered (typecheck+tsc), same WS adapter

.hermes/plans/2026-08-18_gmw-public-features-7-15.md

@@ -0,0 +1,119 @@
+# GMW — Fitur Publik Lanjutan #7–#15 + Bug Fix Reputation Removal
+
+> **For Hermes:** Implement task-by-task. Build + lint + typecheck each service after its
+> changes. Deploy via push to main (CI handles Nix build + systemd). Apply any new
+> drizzle migration MANUALLY (systemd does NOT run migrations).
+> Hard constraint (user): public read-only web, fully automatic, rules in code,
+> NO admin endpoints, NO shadow mode, NO per-user reputation aggregation.
+
+## Bug fix discovered during planning (MUST do first)
+`services/backend/src/modules/dashboard/dashboard.repository.ts` still references
+`pgUserReputationsTable` (import line 8; JOINs at lines 173 + 457) — that table was
+DROPPED in migration `0016`. `dashboard.listUsers` / `dashboard.userDetail` will
+**crash at runtime** (undefined table). Remove the import + the `r.*` join columns
+(`trust_score`, `clean_message_streak`, `total_infractions`) from both queries.
+This is a regression introduced by the reputation removal commit.
+
+## Features to implement (#7–#15)
+All reuse existing infra: `moderation_actions`, `messages`, `channel_cultures`,
+`term_glossary_cache`, `ai_analysis_runs`, `message_edits`, gateway cron (for #15),
+WS (proven Live Feed pattern), oRPC over WS (proven), pure-SVG charts (no libs).
+
+| # | Feature | Data source | Surface |
+|---|---------|-------------|---------|
+| 7 | Flagged Link / Scam Domain Reporter | regex URL from `moderation_actions.content`/`evidence` | `/moderation` |
+| 8 | Top Flagged Channels | join `moderation_actions.message_id`→`messages.channel_id` | `/moderation` |
+| 9 | Moderation Heatmap by Hour | `moderation_actions.created_at` hour-of-day | `/moderation` |
+| 10 | Flag Category Drill-down | `moderation_actions.categories` (reuse Trends) | `/moderation` FE-only |
+| 11 | Channel Culture Glossary | `channel_cultures` (exists) | new `/channels` panel |
+| 12 | Term Knowledge Base | `term_glossary_cache` (exists) | new `/glossary` panel |
+| 13 | Edit/Evasion Tracker | `message_edits` (exists) | `/messages` |
+| 14 | Auto-mod Coverage Stats | `ai_analysis_runs` (exists) | `/moderation` metric tiles |
+| 15 | Weekly Digest (auto, cron) | aggregate #7/#8/#9 → Discord via gateway cron | gateway cron + `/moderation` |
+
+## Architecture per layer
+
+### Backend (oRPC, `services/backend/src`)
+- New repository methods (add to existing repos, follow `getTrends` SQL style):
+  - `moderation.repository.ts`:
+    - `getTopFlaggedDomains(days)` — `regexp_matches(content,'https?://([^/\s]+)')` on
+      `moderation_actions WHERE created_at>=since`, group by host, COUNT, order DESC LIMIT 20.
+    - `getTopFlaggedChannels(days)` — join `moderation_actions a` LEFT JOIN `messages m`
+      ON `m.id=a.message_id`, group by `m.channel_id`, COUNT, order DESC LIMIT 15.
+      Channel name via `m.metadata::jsonb->'channel'->>'channelName'`.
+    - `getHourlyModeration(days)` — `EXTRACT(HOUR FROM to_timestamp(created_at/1000))`
+      group by hour, COUNT, severity breakdown. (24 rows)
+    - `getFlaggedByCategory(days, category)` — list actions where `categories` contains
+      `category` (reuse `listActions` filter or new query), for drill-down #10.
+    - `getCoverage(days)` — from `ai_analysis_runs`: total runs, status breakdown
+      (clean/flagged/warn/error/pending), coverage % = (analyzed)/(captured in window).
+  - `dashboard.repository.ts` (or new `knowledge.repository.ts`):
+    - `listChannelCultures(limit, search?)` — `channel_cultures` rows (channel_id,
+      guild_id, channel_name from messages metadata, culture_summary, last_analyzed_at).
+    - `listGlossary(limit, search?)` — `term_glossary_cache` (term, definition, source_url,
+      resolved_at, hit_count) order by hit_count DESC.
+  - `messages.repository.ts`:
+    - `getEditHistory(limit, channelId?)` — `message_edits` join `messages` for
+      old_content + channel + username + edited_at, order DESC LIMIT.
+- `moderation.service.ts` / `dashboard.service.ts` / `messages.service.ts`: thin wrappers.
+- `orpc/router.ts`: add procedures (follow `trends` shape):
+  - `moderation.topDomains`, `moderation.topChannels`, `moderation.byHour`,
+    `moderation.byCategory` (input `{days,category}`), `moderation.coverage`.
+  - `dashboard.channelCultures`, `dashboard.glossary`.
+  - `messages.editHistory`.
+
+### Frontend (`services/frontend/src`)
+- `lib/types/moderation.ts`: add `FlaggedDomain`, `FlaggedChannel`, `HourlyModeration`,
+  `ModerationCoverage` interfaces.
+- `lib/types/index.ts` (+ message.ts): add `ChannelCultureRow`, `GlossaryRow`, `EditHistoryRow`.
+- `lib/api/moderation.ts`: add `topDomains`, `topChannels`, `byHour`, `byCategory`, `coverage`.
+- `lib/api/dashboard.ts` (or messages.ts): add `channelCultures`, `glossary`, `editHistory`.
+- `lib/api/server.ts`: add SSR seed fetchers (follow `getModerationStats`).
+- `hooks/use-moderation.ts`: add `useTopDomains`, `useTopChannels`, `useHourlyModeration`,
+  `useByCategory`, `useCoverage`. `hooks/use-dashboard.ts`/`use-messages.ts`: add culture/glossary/edit hooks. `hooks/index.ts`: export all.
+- New components (pure SVG/CSS, reuse `GlassPanel`/`SectionHeader`/`Badge`/`Donut`):
+  - `components/ScamDomains.tsx`, `components/TopChannels.tsx`, `components/ModerationHeatmap.tsx`,
+    `components/CoverageTiles.tsx`, `components/ChannelCultureGlossary.tsx`,
+    `components/TermGlossary.tsx`, `components/EditHistory.tsx`.
+- Wire into `app/(dashboard)/moderation/view.tsx` (grid col-span-2/3/5 as space allows)
+  and `app/(dashboard)/messages/view.tsx` (EditHistory panel) and new route pages
+  `app/(dashboard)/channels/page.tsx` + `app/(dashboard)/glossary/page.tsx` with
+  matching `view.tsx` (follow existing page→view SSR pattern; check `app/(dashboard)/dashboard/page.tsx`).
+- Export CSV buttons reuse `lib/csv.ts` `downloadCsv` (client-side) for domains/channels/edits.
+
+### Gateway (#15 Weekly Digest)
+- Add a cron/interval in `services/discord-gateway` (check existing scheduler pattern —
+  search `setInterval`/`cron` in `src`). On a 7-day cadence, query backend oRPC
+  (`dashboard.activity`, `moderation.trends`, `moderation.topChannels`) — OR compute
+  directly via a shared repository — and post a formatted summary to the monitor guild
+  channel (via existing `discordClient.channels.send` helper). Fully automatic, no UI.
+
+## Files touched (summary)
+- backend: `modules/moderation/{repository,service}.ts`, `modules/dashboard/{repository,service}.ts`,
+  `modules/messages/{repository,service}.ts`, `orpc/router.ts`, `shared/index.ts` (if new tables),
+  `lib/types/*` (FE)
+- frontend: `lib/api/*`, `lib/types/*`, `hooks/*`, `components/*`, `app/(dashboard)/*`
+- gateway: new digest scheduler + (none if reuse backend) maybe `shared/redis-channels.ts`
+
+## Constraints / pitfalls (from gmw-ops skill)
+- `created_at` is bigint epoch-MS — compare with `<`/`>`, do NOT divide by 1000 in SQL.
+- Pure SVG only — frontend has ZERO chart libs.
+- `Badge` Tone = signal|amber|vermilion|neutral (no "rose").
+- Frontend WS import is `@/lib/ws/context`; method `on` not `subscribe`.
+- Commit author `asepharyana`, no Co-Authored-By.
+- Rebuild `dist/` after gateway changes; apply drizzle migrations manually.
+
+## Verification
+- Per service: `pnpm typecheck && pnpm lint && pnpm build` green.
+- Gateway: `pnpm test` (117+ pass).
+- Live: `moderation/stats` WS returns data (proves adapter); new procedures registered
+  (typecheck = proof). `systemctl show` new ActiveEnterTimestamp after deploy.
+- DB: confirm `channel_cultures`/`term_glossary_cache`/`message_edits`/`ai_analysis_runs`
+  have rows before relying on them (some may be empty → components handle empty state).
+
+## Execution order
+1. Bug fix dashboard.repository (reputation JOIN) — deploy-safe.
+2. Backend repositories + service + router (#7,#8,#9,#14 dashboard; #11,#12; #13).
+3. FE types + api + hooks + components + wire (#7,#8,#9,#10,#11,#12,#13,#14).
+4. Gateway #15 digest (if scheduler exists) — verify via log, not UI.
+5. Build/lint all 3 services; commit; push; monitor CI; apply migrations; verify live.

.hermes/plans/2026-08-22_154707-ai-analysis-audit-optimize.md

@@ -0,0 +1,111 @@
+# AI Analysis Flow — Audit & Optimization (discord-gateway)
+
+**Goal:** Analisis alur AI analysis end-to-end, temukan bug/inconsistency yang merusak kualitas verdict, lalu perbaiki root cause-nya.
+
+## Scope
+- `services/discord-gateway/src/modules/ai-moderation/**`
+- Tidak menyentuh chatbot backend / frontend.
+
+## Alur saat ini (hasil tracing)
+```
+message capture → aiAnalyzer.queueMessageAnalysis(messageId)
+  → batchScheduler.scheduleConversationAnalysis(conversationKey)   [debounce 250ms, CB gate]
+  → messageStore.getPendingMessagesByConversation(≤200)
+  → skipAgeRestrictedMessages
+  → pickBatchWithinBudget(14000 tokens, 50/msg)
+  → processBatch [Piscina worker, ≤4 threads]
+      → ai-analysis-worker.processBatch
+          → getConversationContextBefore(20 msgs) + attachments
+          → attachment-upload race guard (pending upload → skip)
+          → runModerationAnalysis
+              → Phase 1: exact-hash cache (PG text_analysis_cache, per channel/thread)
+              → Phase 2: semantic cache (embedTexts → Qdrant batch search; PG fallback)
+              → split text-only vs media
+              → runTextOnlyBatch: URL fetch + wiki search + glossary (paralel)
+                  → dedup short messages → sub-batches (60/sub-batch)
+                  → vision evidence utk URL images (hoisted, 15s cap per image)
+                  → callModerationLLM per sub-batch (stream:true, retries 3, JSON parse + correction retry)
+              → runMediaBatch: download → vision per image (cache LRU→DB→live, lock) → 1 LLM call
+              → setCachedTextModeration (PG + Qdrant upsert w/ embedding)
+          → normalizeResult (confidence clamp, fallback analysis)
+          → updateMessagesAIAnalysisBulk → broadcast + scheduleAutoDelete
+  → recovery worker tiap 10s: pending keys → re-schedule; incomplete → individual fallback queue
+  → individual fallback: 1 msg = 1 worker job (context + full LLM)
+  → cache prune tiap 6 jam (PG expired + Qdrant expired points)
+```
+
+## Temuan audit (ranked)
+
+### F1 — Cache hit menghapus status "warn" (BUG AKURASI)
+`moderationOrchestrator.ts` Phase-2 semantic hit & PG-fallback memetakan status via
+`parseQdrantVerdict`: storedStatus bukan "warn"/"flagged" → dipaksa "clean".
+TAPI exact-hash lookup (`getCachedTextModeration`, textCacheStore.ts:288-295) lebih parah:
+hanya menerima "clean"|"flagged" — **"warn" jatuh ke branch flags.length===0 ? clean : flagged**
+→ warn dengan flags=["conflict_instigation"] dibaca sebagai FLAGGED.
+Efek: auto-delete eligibility (butuh recommendedAction delete/escalate + severity list) salah baca;
+dashboard menampilkan flagged padahal verdict asli warn. Root cause: type narrowing legacy
+(`status: "clean" | "flagged"`) tidak diupdate ketika "warn" ditambahkan ke schema.
+
+### F2 — Exact-cache key mengabaikan edit (BUG EVASION)
+Key = sha256(content)+context. Pesan yang DIEDIT (`edited_content`) menghasilkan hash berbeda,
+tapi verdict lama utk konten pre-edit tetap hidup; lebih penting: pesan edited="true" adalah sinyal
+evasion di prompt, sedangkan cache bisa menyajikan verdict dari konten lama jika content sama.
+(Minor, tapi konsistensi: `resolveIsEdited` ada di prompt, tidak ada di cache key.)
+
+### F3 — `pickBatchWithinBudget` skip-bukan-break (LATENSI/KUALITAS)
+Loop `if (usedTokens + msgTokens <= maxTokens) {push}` — pesan BESAR di tengah list dilewati
+dan iterasi lanjut mencoba msg berikutnya. Efek: batch berisi "lubang" (msg pending tetap pending,
+dianalisis di gelombang berikutnya = LLM call tambahan). Ini by-design tolerable, tapi ada bug halus:
+pesan >budget tunggal tidak pernah masuk (scheduler sudah punya fallback slice(0,1), OK).
+Keputusan: biarkan (bukan bug nyata), catat saja.
+
+### F4 — `callModerationLLM` max_tokens 16384 hardcoded (COST)
+Sub-batch 60 pesan × output ~150 token/pesan ≈ 9k token cukup; 16k aman. Biarkan.
+
+### F5 — Dead code builder user-profile/reputation
+`buildUserProfilesBlock`, `buildUserProfileRef`, `UserProfileEntry` di moderationBuilders.ts
+tidak dipakai lagi sejak context minimization (hanya tests). `<user_history>` juga tak pernah
+di-inject (rules masih menyebutnya — misleading bagi model). Bersihkan referensi prompt.
+
+### F6 — rules.ts menyebut `<user_history>` yang tidak pernah ada di payload
+Model diberi instruksi tentang blok yang tak pernah muncul → pemborosan token + potensi
+kelakuan aneh ("menunggu" data yang tak ada). Hapus/ubah kalimat.
+
+### F7 — system.ts "Blok Data" menyebut `<term_glossary> (SearXNG)` — STALE
+Sumber sudah Wikipedia. Komentar kode & teks prompt menyebut SearXNG. Perbaiki teks (kecil).
+
+### F8 — output.ts typo "secifik", baris tabel `-|-` rusak
+Kualitas prompt: typo + markdown table broken (`||-`) di beberapa baris. Rapikan.
+
+### F9 — llmCaller parse-error correction tail hanya di SYSTEM
+Correction tail ditambahkan ke system prompt; provider caching fine, tapi preview invalid
+content (800 char) ikut SYSTEM — ok. Skip.
+
+### F10 — `getLlmSemaphore` race kecil saat config berubah di tengah flight
+Non-issue praktis (config statis per proses). Skip.
+
+## Keputusan perbaikan (yang dieksekusi sekarang)
+1. **F1 (utama):** normalisasi status di SATU tempat — `normalizeStoredStatus()` di
+   textCacheStore.ts yang menerima clean/warn/flagged; pakai di getCachedTextModeration
+   DAN parseQdrantVerdict; perluas return types ke union penuh. Orchestrator tinggal pakai.
+2. **F6+F7+F8:** bersihkan stale references di prompts (user_history, SearXNG, typo).
+3. **F5:** hapus dead builders + test-nya (biome/tsc yang jaga).
+4. Regression test untuk F1 (vitest): warn tersimpan → warn terbaca (exact + qdrant path).
+
+## Files touched
+- services/discord-gateway/src/modules/ai-moderation/textCacheStore.ts (F1)
+- services/discord-gateway/src/modules/ai-moderation/moderationOrchestrator.ts (type only)
+- services/discord-gateway/src/modules/ai-moderation/prompts/rules.ts (F6)
+- services/discord-gateway/src/modules/ai-moderation/prompts/system.ts (F7)
+- services/discord-gateway/src/modules/ai-moderation/prompts/output.ts (F8)
+- services/discord-gateway/src/modules/ai-moderation/moderationBuilders.ts (F5)
+- services/discord-gateway/tests/contextEnrichment.test.ts (F5 test cleanup + F1 regression test baru)
+
+## Verification
+```
+cd services/discord-gateway
+npx tsc --noEmit
+npx biome check --diagnostic-level=error .
+npx vitest run
+```
+Semua harus hijau sebelum commit. Deploy via GHA (push main) — user konfirmasi belakangan.

.hermes/plans/2026-08-22_nonissue-optimizations.md

@@ -0,0 +1,33 @@
+# Optimisasi "non-issue" AI analysis pipeline
+
+## Scope
+Dua item yang sebelumnya dinyatakan non-issue, kini dioptimalkan + 1 bug ordering
+yang ditemukan saat menelusuri:
+
+1. **pickBatchWithinBudget: skip → break.** Pesan diurutkan `created_at ASC`
+   oleh DB. Setelah budget habis, pesan berikutnya pasti lebih besar/lebih kecil
+   arbitrer — skip-then-take menghasilkan batch non-kontigu (ada gap analisis
+   di tengah timeline). Ubah jadi stop at first overflow (break) supaya prefix
+   kronologis utuh; sisanya otomatis diambil gelombang berikutnya
+   (`shouldScheduleNext` sudah selalu true setelah sukses).
+2. **max_tokens dinamis.** Hard-coded 16384 di llmCaller.ts → parameter
+   opsional `maxTokens?`; default tetap 16384. Caller text/media batch pass
+   nilai berbasis ukuran prompt (tiktoken) dengan floor/ceiling.
+3. **Bug ordering UPDATE..RETURNING (bonus).** messagesAnalysis.ts
+   `getPendingMessagesByConversation`: SELECT ids di-order `created_at ASC`
+   tapi UPDATE...RETURNING tanpa ORDER BY → urutan rows balik tidak
+   terjamin. Konsumen pakai messages[0] sebagai anchor konteks
+   (beforeCreatedAt) dan pickBatchWithinBudget asumsi urutan. Fix: re-sort in
+   JS by created_at (stable) sebelum return.
+
+## Files touched
+- src/modules/ai-moderation/batchProcessor.ts — break bukan skip; test baru.
+- src/modules/ai-moderation/llmCaller.ts — param maxTokens.
+- src/modules/ai-moderation/textBatchProcessor.ts / mediaBatchProcessor.ts —
+  hitung token prompt & pass maxTokens.
+- src/modules/message-capture/messagesAnalysis.ts — sort hasil RETURNING.
+- tests/batchBudget.test.ts — baru.
+
+## Verification
+cd services/discord-gateway && bun run typecheck && bun run lint && bun run test
+lalu commit+push, watch GHA, restart service via deploy pipeline.

.hermes/plans/2026-08-22_voice-audio-playback-polish.md

@@ -0,0 +1,67 @@
+# Spec: Perbagus fitur Voice + Audio Playback (GMW frontend)
+
+Tanggal: 2026-08-22 · Scope: **frontend only** (backend/gateway API sudah cukup)
+
+## Masalah (audit)
+1. Recordings: semua kartu pakai `<audio controls>` native — tampilan identik,
+   tidak ada indikasi which-clip-playing / loading / paused, dan N audio bisa
+   play bareng (overlap).
+2. Media view: `thumbnailUrl` dari gateway tidak dipakai; tidak ada visual
+   "sedang playing" selain disc spin; queue item semua sama tanpa badge up-next.
+3. Mini-player (`lib/hooks/use-media-player.tsx`) ada tapi TIDAK PERNAH
+   dimount → dead code, user tidak lihat status musik di halaman lain.
+4. Voice page: `useMicTransmit.setVolume` + `useVoiceListen.setVolume`
+   tersedia tapi tak ada UI-nya; mic live tidak punya level feedback.
+
+## Desain
+
+### A. RecordingAudioPlayer (baru, `components/voice/recording-audio-player.tsx`)
+Custom player menggantikan `<audio controls>`:
+- Play/pause button (ikon berubah), spinner saat buffering (`waiting` event).
+- Progress bar seekable (click-to-seek) + time label `m:ss / m:ss`.
+- Waveform-ish equalizer bars saat playing (CSS animation, reduced-motion safe).
+- **Single-playback**: module-level registry `activePlayers` — memainkan satu
+  clip otomatis pause yang lain.
+- Kartu pemilik player aktif dapat highlight border signal + "Now playing" chip.
+
+### B. Recordings view — pasang player baru
+- Ganti `<audio>` → `<RecordingAudioPlayer src download_url>`.
+- Highlight kartu via state lifted: `playingId` di view, callback `onPlay`.
+
+### C. Media view polish
+- Hero: thumbnail (jika `current.thumbnailUrl`) sebagai disc center image;
+  fallback ListMusic icon. Equalizer bars animasi CSS saat `playing`.
+- Queue row pertama: badge "up next"; baris current track diberi ring signal.
+- Volume read-only tetap.
+
+### D. MiniPlayer global
+- Hapus `lib/hooks/use-media-player.tsx` (dead) — ganti dengan komponen
+  `components/media/mini-player.tsx` yang subscribe `useMediaState` +
+  `useMediaWsSync` langsung (SWR cache shared antar route), mounted di
+  `AppFrame` bawah layar (fixed bottom, hidden di route `/media`).
+- Menampilkan: thumbnail kecil/judul, tombol skip/stop, link ke /media.
+
+### E. Voice UI
+- Mic live: level meter (Equalizer bars) — mic-transmitter sudah punya worklet;
+  tambah `getLevel()` via AnalyserNode pada stream (simple RMS) di hook.
+- Listen: volume slider (input range) wired ke `listen.setVolume`.
+- Mic volume slider wired ke `mic.setVolume`.
+
+## File touched
+| File | Aksi |
+|---|---|
+| services/frontend/src/components/voice/recording-audio-player.tsx | new |
+| services/frontend/src/app/(dashboard)/recordings/view.tsx | edit |
+| services/frontend/src/app/(dashboard)/media/view.tsx | edit |
+| services/frontend/src/components/media/mini-player.tsx | new |
+| services/frontend/src/components/shell/ambient-app.tsx | mount MiniPlayer |
+| services/frontend/src/lib/hooks/use-media-player.tsx | delete |
+| services/frontend/src/hooks/use-voice.ts | tambah micLevel |
+| services/frontend/src/lib/audio/mic-transmit.ts | expose analyser level |
+| services/frontend/src/app/(dashboard)/voice/view.tsx | sliders + meter |
+
+## Verifikasi
+1. `pnpm lint` (biome) + `pnpm build` clean.
+2. Smoke di port **4024** (BUKAN 4017) → curl 200 semua route.
+3. Commit (tanpa trailer) → push → `gh run watch` → live check
+   https://imphnen.asepharyana.my.id/{media,recordings,voice}/ = 200.

.hermes/plans/2026-08-24-ai-analysis-cache-optimization.md

@@ -0,0 +1,127 @@
+# Spec: Optimasi AI Analysis GMW — Naikkan Cache Hit Tanpa Kehilangan Akurasi
+
+Tanggal: 2026-08-24 · Repo: `~/GMW` (branch `main`) · Service: `services/discord-gateway`
+
+## Latar & Evidence (audit 2026-08-24)
+
+State produksi:
+- Qdrant `gmw_text_moderation`: **1.550 poin, status green** (vectors size 2048, Cosine).
+- PG `text_analysis_cache`: 1.634 row `user_moderation`, 277 `vision_llm`; **sum(hit_count) = 0** →
+  hit-rate tidak pernah terukur.
+- Embedding aktif (`AI_LLM_EMBEDDING_MODEL` set, Nemotron-embed, dim 2048), `AI_LLM_EMBEDDING_MIN_SIMILARITY`
+  tidak diset di BWS → default **0.97** (sangat konservatif).
+- Messages: 9.375 total; 643 status `error` (banyak retry), 49 pending.
+
+Temuan audit alur (`moderationOrchestrator.ts` → `textCacheStore.ts` → `qdrantClient.ts`,
+`textBatchProcessor.ts`, `urlFetcher.ts`, `wikipediaClient.ts`, `visionAnalyzer.ts`):
+
+| # | Temuan | Dampak |
+|---|--------|--------|
+| F1 | Exact-hash cache key menyertakan context (channel/thread) → teks sama di channel lain selalu miss | Killer hit-rate #1 |
+| F2 | Semantic tier TIDAK memfilter context (Qdrant payload tak punya context) — sudah global tapi hanya aman krn sim 0.97 ketat | Inkonsisten dgn exact tier |
+| F3 | Phase-1 lookup loop `await getCachedTextModeration(key)` per pesan → N round-trip PgBouncer per batch (60 msg = 60 query serial) | Latensi + beban DB |
+| F4 | Verdict actionable (flagged/warn) dan clean sama-sama boleh di-serve semantic; toleransi akurasi beda | Risiko akurasi |
+| F5 | `hit_count` tidak pernah di-increment oleh reader manapun | Hit-rate tak terukur |
+| F6 | `wikipediaSearch()` (blok `<web_searches>`) tanpa cache — re-fetch tiap batch utk query sama | Latensi + spam ke WP |
+| F7 | `fetchUrlSafely()` tanpa cache — link sama di batch berikutnya di-download lagi penuh | Latensi + bandwidth |
+| F8 | Vision cache key dari data-URL base64 hasil resize → attachment sama via jalur berbeda (URL vs embed) = key beda → re-download + re-vision | Duplikasi kerja vision |
+
+Non-goals: mengubah pipeline enforcement (auto-mute/ban trust-store writes), mengubah prompt
+kebijakan moderasi, mengubah model/embedding provider.
+
+## Desain
+
+Semua perubahan degrade gracefully — cache gagal → perilaku lama (LLM). Akurasi dilindungi
+asimetris: **hemat boleh untuk verdict non-actionable, konservatif untuk yang memicu aksi.**
+
+### D1 — Cache metrics (F5)
+- `textCacheStore.getCachedTextModeration()`: saat hit valid, increment `hit_count`
+  (`UPDATE ... SET hit_count = hit_count + 1`) fire-and-forget (`.catch(()=>{})`), jangan blokir return.
+- Log info periodik ringkas di orchestrator sudah ada ("User moderation cache applied") — cukup.
+
+### D2 — Batched exact-cache lookup (F3)
+- Fungsi baru `getCachedTextModerations(keys: string[]): Promise<Map<string, StoredModerationVerdict>>`
+  di `textCacheStore.ts`: **satu** `SELECT ... WHERE text = ANY($1)` (chunk 200 key/query),
+  parse + `normalizeStoredStatus` per row (reuse helper existing).
+- Orchestrator fase-1: kumpulkan semua key unik → satu call batched → distribusi hasil.
+- Semantik identik dengan loop lama (row expired/error-artifact tetap miss); hanya jumlah round-trip
+  yang turun N→1.
+
+### D3 — Global exact reuse untuk verdict non-actionable (F1)
+- Key scoped-context TETAP ditulis (kompatibel, invalidasi moderator tetap presisi).
+- Reader tambahan: kalau key `<ctx>:<hash>` miss, coba key legacy global `text_mod:<hash>` (bare).
+- Guard akurasi (WAJIB semua terpenuhi):
+  - `status === "clean"` DAN `flags.length === 0`;
+  - `confidence >= AI_CACHE_GLOBAL_REUSE_MIN_CONFIDENCE` (default 0.85);
+  - `recommendedAction === "none"`;
+  - umur entry ≤ `AI_CACHE_GLOBAL_REUSE_MAX_AGE_H` (default 72h) — cek `analyzed_at`.
+- Flag baru `policyVersion: "cached-global-clean-2026-08"` supaya terlacak di dashboard/log.
+- Verdict flagged/warn TETAP context-scoped (tidak pernah lintas channel).
+
+### D4 — Semantic dua-band similarity (F2+F4)
+- Config baru: `AI_LLM_EMBEDDING_MIN_SIMILARITY_ACTIONABLE` default **0.97** (perilaku lama),
+  `AI_LLM_EMBEDDING_MIN_SIMILARITY_CLEAN` default **0.92**, keduanya coerce number 0..1.
+- Satu Qdrant batch search pakai threshold RENDAH (0.92). Per hit, klasifikasi ulang:
+  - verdict non-actionable (clean, no flags, action=none): terima jika `score >= CLEAN_BAND`;
+  - verdict actionable (warn/flagged atau flags ada / action != none): terima hanya jika
+    `score >= ACTIONABLE_BAND` (0.97 — persis gate lama);
+  - di antara dua band → buang hit, pesan lanjut ke LLM (fail-open ke akurasi).
+- Legacy PG fallback path: filter serupa di `findSimilarTextModeration` via parameter band.
+
+### D5 — Cache Wikipedia search (F6)
+- `wikipediaClient.wikipediaSearch(query)`: cek `cacheGet(makeCacheKey("wikisearch", q))` dulu;
+  miss → fetch (timeout existing) → sukses & hasil non-kosong → `cacheSet(..., TTL 6h)`.
+  Hasil kosong TIDAK di-cache (biar retry nanti). Redis down → langsung fetch (no-op cache).
+
+### D6 — Cache URL text fetch (F7)
+- `urlFetcher.fetchUrlSafely(url)`: wrapper async memoize in-process LRU (max 500, TTL 30 menit)
+  untuk `type === "text"` saja (image tetap selalu fresh-download karena dipakai sbg bukti vision
+  + buffer besar; error tidak di-cache).
+- Import `LRUCache` dari `lru-cache` (sudah dep gateway).
+
+### D7 — Unified vision cache key (F8)
+- `makeImageCacheKey(imageUrl)` di `textCacheStore.ts`: sebelum hash, strip query Discord CDN
+  (`?ex=&is=&hm=` signed tokens, `format/width/height/size`) — regex `(\?[^#]*)$` dibuang bila host
+  CDN discord (`cdn.discordapp.com`, `media.discordapp.net`, `images-ext-*.discordapp.net`);
+  URL non-Discord: hash full URL seperti sekarang.
+- Efek: attachment sama yang lolos lewat jalur embed vs inline vs re-fetch dgn token beda → SATU
+  entry cache → skip download+vision kedua kali. Data-URL base64 tetap di-hash apa adanya.
+
+## File yang disentuh
+
+1. `src/shared/config/index.ts` — 3 config baru (D3×2, D4×2 — total 4 nilai, 3 baris zod + deskripsi).
+2. `src/modules/ai-moderation/textCacheStore.ts` — hit_count inc (D1), batched getter (D2),
+   global-reuse guard helper (D3), image-key normalize (D7).
+3. `src/modules/ai-moderation/moderationOrchestrator.ts` — pakai batched getter (D2),
+   global bare-key fallback (D3), dua-band semantic accept (D4).
+4. `src/modules/ai-moderation/qdrantClient.ts` — `searchQdrantBatch` menerima threshold rendah
+   (sudah parametrik — mungkin tanpa perubahan; verifikasi).
+5. `src/modules/ai-moderation/wikipediaClient.ts` — cache layer (D5).
+6. `src/modules/ai-moderation/urlFetcher.ts` — LRU text-fetch memoize (D6).
+
+## Schema/type changes
+
+- Tidak ada migrasi DB (kolom `hit_count`, `analyzed_at`, `expires_at` sudah ada).
+- Tidak ada perubahan kontrak WS/oRPC/frontend.
+- Type baru: none public; internal `StoredModerationVerdict` dipakai ulang.
+
+## Verification
+
+1. Unit tests baru (`tests/`):
+   - `cacheBatchLookup.test.ts`: batched getter — hit/miss/expired/error-artifact mapping,
+     chunking >200 keys (mock executeAll), hit_count increment called.
+   - `globalReuseGuard.test.ts`: guard menerima clean+conf≥0.85+action none+umur ≤72h;
+     menolak flagged/warn/conf rendah/action≠none/stale.
+   - `semanticBands.test.ts`: clean @0.93 diterima, flagged @0.93 ditolak, flagged @0.98 diterima.
+   - `imageKeyNormalize.test.ts`: URL Discord dgn/ex token → key sama; non-Discord beda query → beda.
+2. Gate service: `pnpm typecheck && pnpm exec biome check --diagnostic-level=error . && pnpm exec vitest run`.
+3. Deploy via GHA (`git push origin main`) → watch `Build & Deploy (Nix)` → verifikasi
+   `systemctl show gmw-discord-gateway -p ActiveEnterTimestamp` baru.
+4. Runtime probe pasca-deploy: journalctl level 30 normal; beberapa jam kemudian
+   `SELECT sum(hit_count) FROM text_analysis_cache WHERE source='user_moderation'` > 0 membuktikan
+   metrics jalan; log "User moderation cache applied" menunjukkan hits>0 pada traffic ramai.
+
+## Rollback
+
+Semua fitur behind config defaults yang mempertahankan perilaku lama pada nilai konservatif;
+rollback = redeploy commit sebelumnya (tanpa migrasi DB, tanpa state eksternal).

.hermes/plans/2026-08-24-attachment-delay-fix.md

@@ -0,0 +1,56 @@
+# Spec: Perbaiki Delay Attachment 162s→<20s (GMW AI Analysis)
+
+Tanggal: 2026-08-24 · Repo `~/GMW` · Service discord-gateway
+
+## Evidence (audit produksi)
+
+Klaster pesan attachment delay ~330–400 detik. Trace pesan `1541417073245290638` (.gif):
+19:01:08 dibuat → 19:01:09 batch incomplete → fan-out individual → **guard upload-pending
+mengembalikan `results:[]`** → diperalakukan sukses (`complete ... (undefined)`) → row
+tertahan `ai_status='processing'` **tanpa penanggung jawab** → 19:06:12 cleanup mengembalikan
+ke `pending` (tepat 300s) → baru dianalisis. Plus vision gagal 3× utk GIF besar
+("Stream ended before producing a non-ping SSE event") → degradasi teks.
+
+## Root causes
+
+- **A (fatal)**: `individualFallbackProcessor.processIndividualFallback` memperlakukan
+  `ok:true + results:[]` sebagai sukses. Race-guard upload di `ai-analysis-worker.processIndividual`
+  sengaja balik `results:[]` (desain lama) → pesan yatim `processing` sampai cleanup 300s.
+- **B**: `llmVision` hanya mencoba `stream:true`; kegagalan SSE truncation pada gambar besar
+  = 3 retry sia-sia (semua jalur sama) → bukti media hilang.
+- **C**: safety-net cleanup 300s terlalu lambat sbg satu-satunya pemulih `processing`.
+
+## Fix
+
+1. **F1 — sinyal eksplisit upload-pending**: `IndividualOkResponse` + field opsional
+   `uploadPending?: boolean`. Worker set `uploadPending:true` saat race guard kena.
+2. **F2 — processor menangani 3 kondisi** via helper murni baru
+   `classifyIndividualWorkerResult(result): "success" | "upload_pending" | "incomplete" | "error"`
+   (modul baru `fallbackResultClassifier.ts`, zero-dep agar mudah dites):
+   - `upload_pending` → tulis ulang row ke `pending` (pola sama dgn revert apiFailed di
+     batchProcessor) + broadcast + **re-schedule analisis percakapan segera**
+     (dynamic import batchScheduler, pola anti-siklus yg sudah ada) → retry dalam ~250ms
+     begitu upload beres. Bukan error, tidak naikkan CB counter.
+   - `incomplete` (flags analysis_incomplete) → perilaku lama (exhausted path).
+   - `error` / `results kosong tanpa penjelasan` → throw transien (retry oleh recovery),
+     BUKAN sukses palsu. Log "(undefined)" hilang.
+3. **F3 — vision non-stream fallback**: di `llmVision`, jika error match
+   `/Stream ended before producing a non-ping SSE|stream ended/i` → coba SEKALI lagi dengan
+   `stream:false` (router agregasi penuh; timeout tetap 60s). Konversi hard-fail jadi sukses.
+4. **F4 — turunkan safety net**: default `revertStuckProcessingMessages` 300000 → 120000 ms.
+
+## File disentuh
+
+- `src/modules/ai-moderation/fallbackResultClassifier.ts` (BARU, pure)
+- `src/modules/ai-moderation/ai-analysis-worker.ts` (tipe + set flag uploadPending)
+- `src/modules/ai-moderation/individualFallbackProcessor.ts` (konsumsi classifier + reschedule)
+- `src/modules/ai-moderation/llmClient.ts` (fallback non-stream di llmVision)
+- `src/modules/message-capture/messagesCleanup.ts` (default 120s)
+
+## Verifikasi
+
+- Test baru `tests/fallbackResultClassifier.test.ts` (4 klasifikasi + edge kosong).
+- Gate: tsc --noEmit, biome error-level, vitest run semua hijau.
+- Deploy GHA sukses; pasca-deploy: pesan attachment baru p50 < 20s
+  (`SELECT percentile_cont(0.5) ... WHERE metadata attachments>0 AND created_at > deploy`),
+  tidak ada lagi "complete ... (undefined)".

.hermes/plans/2026-08-24_135602-gmw-fe-constellation-rombak.md

@@ -0,0 +1,205 @@
+# GMW FE Rombak — "Constellation Ops" Implementation Plan
+
+> **For Hermes:** Use subagent-driven-development skill to implement this plan task-by-task.
+> Setiap task = edit → lint → build/test → smoke → commit (author `asepharyana`, tanpa Co-Authored-By).
+
+**Goal:** Rombak total FE GMW dari template dashboard klasik menjadi **Constellation Node Graph** — graf node interaktif (three.js force layout) di mana channel/event/moderasi adalah node bintang dan navigasi adalah terbang antar node — untuk semua 9 route sekaligus.
+
+**Architecture:** Hapus total shell template (TopBar + NavRail + MobileNav + main scroll container). Ganti dengan satu `ConstellationStage`: canvas full-bleed fixed di belakang seluruh app yang me-render force-directed graph per-route (scene), plus overlay HTML mengambang (tanpa card box standar) untuk konten detail. Router tetap Next App Router; perpindahan route = kamera "fly-to" ke scene baru. Data tetap lewat pola SSR seed (`page.tsx` server fetch → `view.tsx` client + SWR fallbackData) Line truncated
+
+**Tech Stack:** Next.js 16 App Router · React 19 · TypeScript strict · Tailwind v4 (token oklch existing) · three r180 (+ `@types/three`) · d3-force (+ `@types/d3-force`) untuk simulasi layout deterministik · motion v12 (transisi antar-scene) · biome (0 error **dan** 0 warning) · bun:test untuk modul murni.
+
+---
+
+## Keputusan Terkunci (2026-08-24, user MythEclipse)
+
+| Keputusan | Pilihan user |
+|---|---|
+| Cakupan | **Semua 9 route sekali jalan** (dashboard, voice, media, recordings, moderation, messages, analysis, channels, glossary) |
+| Arah struktural | **Constellation Node Graph** — navigasi = terbang antar node |
+| Tema | **Pertahankan toggle light/dark** (jangan rebuild sistem tema — harden saja) |
+
+## Kondisi Saat Ini (fakta repo, 2026-08-24)
+
+- Repo: `/home/code/GMW/services/frontend` (monorepo GMW, deploy Nix-first via GHA "Build & Deploy (Nix)" ke `imphnen.asepharyana.my.id`; nginx gmw-proxy :4009 → Next standalone :4017).
+- Shell aktif: `src/components/shell/ambient-app.tsx` → `AppFrame` = NavRail + TopBar + `<main>` scroll + MobileNav + MiniPlayer, dipakai SEMUA route via `(dashboard)/layout.tsx`. AmbientCanvas (233 l) masih ada sebagai background di belakang chrome template.
+- 9 route di `src/app/(dashboard)/`: dashboard (368 l), voice (311), media (293), moderation (424), messages (611), analysis (196), recordings (201), channels (23), glossary (15).
+- `next.config.ts`: `output: "standalone"` + `trailingSlash: true` → **nav via `router.push`/Link bisa no-op; pakai `<a href>` polos** (pitfall terverifikasi).
+- Token desain ada di `globals.css` `@theme`: `--color-canvas/-2`, `--color-ink/-soft/-faint`, `--color-signal(+glow)`, amber, vermilion, hairline; font Bricolage Grotesque/JetBrains Mono; `.light` class override + next-themes sudah jalan — JANGAN dibangun ulang.
+- Sudah ada & dipertahankan: CommandPalette (⌘K), Chatbot, MiniPlayer, primitives (GlassPanel/GlassCard/Badge/…), shared states+skeletons, `lib/format.ts`, `lib/ai-status.ts`, WS typed events (`message_created/updated/deleted/analyzed/snapshot*`, `voice_state`, dll).
+- Tooling lokal: `pnpm lint|build|format`, `pnpm exec biome check --write --unsafe .` buat import order; bun 1.3.14 tersedia untuk unit test modul murni.
+- Smoke test SELALU di port non-prod (mis. 4024), cek `ss -ltnp | grep <port>` dulu; JANGAN pernah bind 4017.
+
+## Struktur Target
+
+```
+src/components/shell/
+  constellation-frame.tsx    # pengganti AppFrame: tanpa topbar/rail/bottom-bar
+  constellation-stage.tsx    # canvas three.js full-bleed + hit-test + kamera
+  route-scenes.ts            # map pathname → scene config (nodes/edges/focus)
+  floating-chrome.tsx        # brand mark + status dot + route switcher minim + palette hint
+  (hapus setelah migrasi): topbar.tsx, nav-rail.tsx, mobile-nav.tsx, ambient-app.tsx*
+src/lib/constellation/
+  graph.ts                   # tipe GraphNode/GraphEdge + builder dari data API (murni)
+  layout.ts                  # wrapper d3-force deterministik (seeded) + fallback radial (murni)
+  camera.ts                  # interpolasi kamera fly-to (murni, testable)
+  *.test.ts                  # bun:test utk modul murni di atas
+```
+
+*ambient-canvas/context dievaluasi: bisa di-refactor jadi bagian dari stage atau dihapus.
+
+### Scene per route (bahasa visual satu kesatuan)
+
+| Route | Scene |
+|---|---|
+| `/dashboard/` | Guild = bintang pusat; channel = node orbit; event WS = pulse edge; metric cluster floating (tanpa box) |
+| `/channels/` | Semua channel sebagai konstelasi; klik channel = fly-to + panel detail mengambang |
+| `/moderation/` | Pesan flagged = node merah (vermilion) mengelilingi hub AI-verdict; LiveModerationFeed jadi ribbon ticker |
+| `/messages/` | Stream pesan = sabuk orbital; node baru masuk lewat animasi dari tepi |
+| `/voice/` | Stage node di tengah; speaker aktif = node mengorbit dengan glow intensitas |
+| `/media/` | Galeri spiral node (thumbnail via div bg-image); klik = MiniPlayer handoff |
+| `/recordings/` | Variasi spiral media, timeline density sebagai cincin |
+| `/analysis/` | Hub analitik; chart existing dirender sebagai panel translusen mengambang ter-anchor ke node |
+| `/glossary/` | Term KB = cincin satelit; search memfilter node real-time |
+
+### Checklist anti-default (wajib lolos sebelum commit akhir)
+
+- [ ] Tidak ada top bar / nav rail / side panel / bottom prompt bar
+- [ ] Tidak ada card grid standar sebagai struktur utama
+- [ ] Layout beda SPASIAL dari versi lama (bukan ganti warna)
+- [ ] Verifikasi visual sungguhan: browser_navigate + browser_vision → "bukan template dashboard"
+
+---
+
+## Tasks
+
+### Phase 0 — Fondasi
+
+#### Task 1: Dependencies + scaffold direktori
+**Files:** `package.json`, create `src/lib/constellation/`
+1. `cd /home/code/GMW/services/frontend && pnpm add three @types/three d3-force @types/d3-force && pnpm add -d @types/bun`
+2. Buat 4 file kosong bertipe sesuai struktur target (ekspor placeholder bertipe agar tsc lolos).
+3. Run: `pnpm exec tsc --noEmit` → expected PASS.
+4. Commit: `chore(frontend): scaffold constellation lib + deps`.
+
+> Catatan: jika `@types/bun` bentrok dengan tsconfig Next (duplikat globals), fallback: exclude `"**/*.test.ts"` dari `tsconfig.json` dan jalankan test hanya via `bun test` (tanpa typecheck). Putuskan saat eksekusi, dokumentasikan di commit body.
+
+#### Task 2 (TDD): Modul murni graph + layout + camera
+**Files:** `src/lib/constellation/{graph,layout,camera}.ts` + `*.test.ts`
+1. **Tulis test dulu** (bun:test):
+   - `graph.test.ts`: builder `dashboardToGraph()` menghasilkan node guild pusat + N channel + edges benar dari fixture `/api/dashboard/stats|channels` shape (lihat AGENTS.md; nama channel di `message.metadata.channelName`).
+   - `layout.test.ts`: `computeLayout(nodes, edges, {seed:42})` DETERMINISTIK (dua pemanggilan = posisi identik), semua posisi finite, radius dalam bound; `radialLayout()` fallback untuk reduced-motion.
+   - `camera.test.ts`: `flyTo(from,to,t)` easing monotonic, t=0→from, t=1→to.
+2. Run: `bun test src/lib/constellation` → expected FAIL (belum diimplementasi).
+3. Implementasi minimal sampai PASS.
+4. Run: `bun test src/lib/constellation && pnpm exec tsc --noEmit` → PASS.
+5. Commit: `feat(frontend): constellation graph/layout/camera pure modules (tested)`.
+
+### Phase 1 — Shell Baru
+
+#### Task 3: `ConstellationStage` (canvas renderer)
+**Files:** `src/components/shell/constellation-stage.tsx`
+- `<canvas>` fixed inset-0 `-z-10`; three.js orthographic 2D-ish render node+edge (sprite/glow shader sederhana — jangan over-engineer; titik + garis + glow cukup).
+- Interaksi: drag = pan, wheel/pinch = zoom, hover = highlight node + tooltip title, click node ber-`href` = navigasi via `<a>` semantics (window.location assign — hindari router.push no-op).
+- DPR-aware resize; pause render loop saat `document.hidden`; `prefers-reduced-motion` → render statis tanpa animasi.
+- Sampling warna dari CSS var (`getComputedStyle`) + `MutationObserver` pada `documentElement.classList` agar ikut toggle dark/light.
+- Verify: `pnpm lint && pnpm build` PASS → commit `feat(frontend): constellation stage canvas renderer`.
+
+#### Task 4: `route-scenes.ts` + `ConstellationFrame`
+**Files:** `src/components/shell/{route-scenes,constellation-frame,floating-chrome}.tsx`
+- `route-scenes.ts`: `usePathname()` → config scene (builder mana, focus node, overlay slots).
+- `ConstellationFrame`: render `ConstellationStage` + `children` (overlay HTML absolute, BUKAN flex column bertingkat) + `floating-chrome` (brand mark kiri-atas dgn status dot WS, switcher route minimal kanan-bawah, hint ⌘K). MiniPlayer + Chatbot + CommandPalette tetap termounting.
+- Switcher & semua nav internal pakai `<a href="/<route>/">` polos (trailingSlash!). Verifikasi klik di browser nanti, bukan cuma baca kode.
+- Wire `(dashboard)/layout.tsx`: ganti `AppFrame` → `ConstellationFrame` (provider Ambient/WS tetap).
+- Smoke: `ss -ltnp | grep 4024` kosong → `PORT=4024 pnpm start` → curl semua 9 route = 200 → kill server.
+- Commit: `feat(frontend): replace classic shell with constellation frame`.
+
+#### Task 5: Hapus chrome lama + bersih-bersih
+**Files:** hapus `shell/topbar.tsx`, `shell/nav-rail.tsx`, `shell/mobile-nav.tsx`, update `shell/index.ts`; eval `ambient-*` (refactor jadi layer stage ATAU hapus).
+- `grep -rn "TopBar\|NavRail\|MobileNav\|AppFrame" src/` → 0 referensi tersisa.
+- `pnpm exec biome check --write --unsafe .` → import order rapi.
+- Commit: `refactor(frontend): remove classic dashboard chrome`.
+
+### Phase 2 — Migrasi 9 Route (batch, tiap batch ship hijau)
+
+Urutan sengaja dari yang paling graph-native ke paling padat konten:
+
+#### Task 6: `/dashboard/` + `/channels/`
+- Dashboard view: guild-star + channel orbit + WS pulse; metric cluster floating (reuse MetricTile tapi tanpa card-box — ubah jadi teks besar + hairline).
+- Channels: konstelasi channel, klik = fly-to + panel detail (data dari hook `use-channels`/server fetch existing — jangan karang endpoint).
+- Lint+build+smoke 2 route → commit `feat(frontend): dashboard & channels constellation scenes`.
+
+#### Task 7: `/glossary/` + `/analysis/`
+- Glossary: cincin satelit term + search filter node (client-side filter pada node label).
+- Analysis: hub node + panel chart translusen mengambang (reuse komponen charts/* existing, anchor posisi ke node).
+- Commit: `feat(frontend): glossary & analysis scenes`.
+
+#### Task 8: `/messages/` + `/moderation/`
+- Messages: sabuk orbital stream (reuse sortMessages/logika WS `use-messages.ts` — jangan duplikasi); pesan baru animate-in dari tepi.
+- Moderation: flagged nodes vermilion + hub verdict; LiveModerationFeed jadi ticker ribbon bawah (bukan panel).
+- Commit: `feat(frontend): messages & moderation scenes`.
+
+#### Task 9: `/voice/` + `/media/` + `/recordings/`
+- Voice: speaker aktif orbit dgn glow (seed dari snapshot `activeSpeakers` server-authoritative — pertahankan perilaku useSpeakers).
+- Media/Recordings: spiral galeri; thumbnail pakai div background-image (hindari `noImgElement`; kalau terpaksa `<img>` → `// biome-ignore lint/performance/noImgElement: <alasan>`); MiniPlayer handoff tidak boleh regress.
+- Commit: `feat(frontend): voice, media & recordings scenes`.
+
+### Phase 3 — Polish, A11y, Tema
+
+#### Task 10: Motion & micro-interaction
+- Fly-to antar scene pakai `motion` v12 (interpolasi kamera dari `lib/constellation/camera.ts`).
+- Overlay: stagger masuk (pattern `animate-stagger` + `staggerDelay(i)` existing); tambah semua `animate-*` baru ke kill-list `prefers-reduced-motion` di `globals.css`.
+- Commit: `feat(frontend): scene transitions + motion polish`.
+
+#### Task 11: A11y & fallback semantik
+- Setiap scene punya fallback semantik: `<nav aria-label="Routes">` visually-hidden berisi link semua route + `<ul>` node (label + href) yang ter-sync dgn graph state.
+- Keyboard: Tab ke daftar route; panah memindahkan fokus antar node; Enter aktivasi; `focus-visible` ring pakai `--color-ring`.
+- Kontras ink-on-canvas dicek (WCAG AA utk teks overlay).
+- Hooks-after-early-return pitfall: compute flags dulu, guard di dalam useEffect.
+- Commit: `feat(frontend): a11y semantic fallback for constellation`.
+
+#### Task 12: Hardening tema light/dark
+- Pastikan canvas sampling var `.light` benar saat toggle (MutationObserver dari Task 3); harden `.light`: `color-scheme: light`, scrollbar/selection, glow opacity turun.
+- JANGAN menyentuh sistem next-themes.
+- Commit: `fix(frontend): light theme parity for constellation stage`.
+
+### Phase 4 — Gerbang Ship (non-negotiable)
+
+#### Task 13: Verifikasi penuh + deploy
+1. `pnpm lint` → **Found 0 errors. Found 0 warnings** (warning pun harus bersih).
+2. `bun test src/lib/constellation` → all PASS; `pnpm exec tsc --noEmit` PASS; `pnpm build` sukses.
+3. Smoke: port 4024 bersih → start → curl 9 route (dgn trailing slash) semua **200** → kill server.
+4. `git status` audit: hanya file scope yang ter-commit (`git checkout --` file liar hasil auto-format).
+5. Push `origin/main` → `gh run watch` "Build & Deploy (Nix)" → hijau.
+6. Live: curl `https://imphnen.asepharyana.my.id/<route>/` × 9 → 200.
+7. **Verifikasi visual sungguhan**: browser_navigate ke live URL → browser_vision: "Apakah ini masih template dashboard?" — jawaban HARUS bukan; screenshot dikirim ke user (`MEDIA:<path>`).
+8. Checklist anti-default dicentang satu per satu di PR/commit message akhir.
+
+---
+
+## Files Likely to Change
+
+```
+package.json, pnpm-lock.yaml
+next.config.ts                        # hanya jika perlu (tidak diharapkan)
+src/app/(dashboard)/layout.tsx        # wire frame baru
+src/app/(dashboard)/{9 route}/page.tsx|view.tsx   # scene migration
+src/components/shell/**               # frame baru + hapus lama
+src/lib/constellation/**              # baru (murni + test)
+src/app/globals.css                   # tokens tambahan + reduced-motion kill list
+biome.json                            # hanya jika rule intentional baru perlu "off"
+```
+
+## Risks / Tradeoffs
+
+| Risiko | Mitigasi |
+|---|---|
+| Perf GPU di device lemah (full-page WebGL) | Node count kecil (<200/route), pause saat tab hidden, fallback radial statis utk reduced-motion; jika tetap berat → downgrade ke Canvas2D renderer dgn API sama |
+| router.push no-op (standalone+trailingSlash) | Semua nav pakai `<a href>`; verifikasi klik nyata di browser |
+| Konten padat (messages 611 l) tak muat di metafora graf | Prinsip: graf = struktur/navigasi; DETAIL tetap lewat overlay panel mengambang — konten tidak dikorbankan, hanya wadahnya yang berubah |
+| Bun test + Next tsconfig konflik types | Exclude `*.test.ts` dari tsconfig (keputusan Task 1) |
+| Regress fitur existing (theme toggle, palette, chatbot, mini-player, WS sync) | Dipertahankan apa adanya; smoke + klik verifikasi per phase; sortMessages & snapshot voice TIDAK disentuh logikanya |
+
+## Open Questions
+
+- Tidak ada blocker. Satu preferensi opsional: apakah brand mark perlu jam/uptime seperti TopBar lama — default: tidak (minimal). Keputusan bisa diambil saat review Task 4.

.hermes/plans/2026-08-24_mono-gamemenu-mobile.md

@@ -0,0 +1,39 @@
+# GMW FE — Monokrom Hitam-Putih + Sidebar Ala Menu Game + Ringan di Mobile
+
+Tanggal: 2026-08-24 · Basis: `eda5c75` (shell usable hasil revert)
+
+## Tujuan
+1. Tema **monokrom murni** (hitam-putih, tanpa warna) di dark & light.
+2. Sidebar (desktop NavRail + mobile dock) beranimasi **ala menu game** — corner
+   brackets, sweep, stagger masuk, marker segitiga.
+3. **Ringan di mobile**: matikan WebGL ambient di layar kecil, kurangi biaya
+   blur/backdrop, animasi transform/opacity saja.
+
+## Non-goals
+- Tidak menyentuh backend, endpoint, hooks/data-flow, struktur route.
+- Tidak menambah dependensi baru (CSS murni untuk semua animasi).
+
+## File yang disentuh
+| File | Perubahan |
+|---|---|
+| `src/app/globals.css` | Token mono (dark+light): signal/amber/vermilion → skala putih-abu; `.glass` blur adaptif; kelas baru `.game-nav-item` (bracket ::before/::after, sweep, stagger via `--i`), `.game-frame` (panel sudut terpotong + garis tergambar), keyframes `sweep-x`, `draw-line`, `nav-in`; media query `<md`: blur 18→8px, hambat animasi berat |
+| `src/components/shell/nav-rail.tsx` | Item pakai `.game-nav-item` + `style={{'--i': n}}`; marker aktif jadi segitiga ▸ putih; hapus box-shadow glow besar (ganti sweep) |
+| `src/components/shell/mobile-nav.tsx` | Dock mono: tab aktif = bar atas putih + sweep sekali; target sentuh ≥44px; hapus glow blob |
+| `src/components/shell/topbar.tsx` | Aksen mono + `.game-frame` pada container (cek markup dulu) |
+| `src/components/ambient/ambient-canvas.tsx` | Early-return WebGL bila `(pointer: coarse)` / lebar <768 / `saveData` / core ≤4; fallback statik CSS tetap |
+| `src/components/ambient/status/signal tone` (`SIGNAL_RGB`) | Semua tone jadi grayscale (putih; intensitas beda per tone) |
+| `src/app/(dashboard)/dashboard/view.tsx` | Hero + kartu metrik pakai `.game-frame`/cut-corner sebagai showcase |
+
+## Keputusan desain
+- **Full monokrom termasuk danger**: flag/moderation tidak lagi merah —
+  ditandai badge putih-di-atlas-hitam inversi + pulse. Kalau user kangen merah,
+  tinggal isi ulang `--color-vermilion`.
+- Semua animasi hanya `transform`/`opacity` (compositor-friendly), hormati
+  `prefers-reduced-motion` (sudah ada kill-switch global).
+
+## Verifikasi (gerbang)
+1. `tsc --noEmit` bersih; biome 0 error 0 warning.
+2. `pnpm build` sukses; smoke lokal 4024 → 9 route 200.
+3. Push → GHA "Build & Deploy (Nix)" hijau → live 9×200.
+4. Visual check live: desktop (rail game-menu terlihat) + cek rule mobile
+   (media query & gate kode) — screenshot disimpan.

flake.nix

@@ -11,11 +11,6 @@
       let
         pkgs = import nixpkgs { inherit system; };
 
-        # libdatachannel for the GoLive N-API binding. nixpkgs 0.24.1 is built
-        # against this host's glibc and ships both lib + dev headers, so the
-        # binding links cleanly inside the Nix sandbox (no manual cmake build).
-        libdatachannel = pkgs.libdatachannel;
-
         # Source filter: `path:` literals do NOT respect .gitignore by default,
         # so a dirty local out/ (stale chunks from previous builds) leaks into
         # the sandbox. Filter out build artifacts explicitly.
@@ -53,8 +48,8 @@
           export GIT_SSL_CAINFO=${pkgs.cacert}/etc/ssl/certs/ca-bundle.crt
           export NIX_SSL_CERT_FILE=${pkgs.cacert}/etc/ssl/certs/ca-bundle.crt
 
-          # pnpm uses node-gyp for native addons — provide build tools
+          # pnpm uses node-gyp for native addons — provide build tools (kept for
-          export npm_config_build_from_source=true
+          # the rare case a prebuilt is unavailable and it falls back to compile).
           export CPPFLAGS="-I${pkgs.lib.getDev pkgs.openssl}/include"
           export LDFLAGS="-L${pkgs.lib.getLib pkgs.openssl}/lib"
 
@@ -73,7 +68,7 @@
         # NOTE: do NOT use `pnpm install --prod` here — it collapses the
         # public-hoist dir (.pnpm/node_modules) that runtime peer resolution
         # relies on (e.g. @lng2004/node-datachannel and @seydx/node-av-linux-x64
-        # are only reachable through it), silently breaking voice/screenshare.
+        # are only reachable through it), silently breaking voice.
         # Instead we keep the full install's symlink layout and only prune
         # orphaned package dirs + broken symlinks.
         # Must run AFTER tsc (typescript is a devDep) and after native builds.
@@ -106,31 +101,8 @@
           buildPhase = pnpmInstall + ''
             echo "=== Compiling TypeScript ==="
             npx tsc 2>&1
-            echo "=== Fixing @/ path aliases to relative paths ==="
+            echo "=== Fixing @/ path aliases + extensionless relative imports for node ESM ==="
-            node -e "
+            node scripts/fix-imports.mjs
-              const fs = require('fs');
-              const path = require('path');
-              let count = 0;
-              function walk(dir) {
-                if (!fs.existsSync(dir)) return;
-                for (const e of fs.readdirSync(dir, {withFileTypes: true})) {
-                  const p = path.join(dir, e.name);
-                  if (e.isDirectory()) walk(p);
-                  else if (e.name.endsWith('.js')) {
-                    const c = fs.readFileSync(p, 'utf8');
-                    const pat = /from\s+['\"]@\/([^'\"]+)['\"]/g;
-                    const n = c.replace(pat, (m, p1) => {
-                      const target = path.join('dist', p1) + '.js';
-                      const rel = path.relative(path.dirname(p), target);
-                      return 'from \"' + (rel.startsWith('.') ? rel : './' + rel) + '\"';
-                    });
-                    if (n !== c) { fs.writeFileSync(p, n); count++; }
-                  }
-                }
-              }
-              walk('dist');
-              console.log('Fixed ' + count + ' files');
-            "
             echo "=== Build complete ==="
           '' + pruneProd;
 
@@ -167,8 +139,7 @@ WRAPPER
             pkgs.pkg-config
             pkgs.openssl
             pkgs.openssl.dev
-            libdatachannel.dev # rtc/rtc.hpp headers for the GoLive binding
+            pkgs.git # for any FetchContent-based deps during native builds
-            pkgs.git # libdatachannel FetchContent clones from GitHub
             pkgs.cacert
           ];
 
@@ -181,65 +152,35 @@ WRAPPER
           # do NOT let stdenv run its own cmake configure phase on the source.
           dontUseCmakeConfigure = true;
 
+          # The gateway bundles native node_modules (.node addons plus .o/.a
+          # object files left in prebuilt dirs). stdenv's fixupPhase walks
+          # $out/node_modules and runs patchELF + shrinkELF over every ELF it
+          # finds, choking on the non-ET_DYN files (.o/.a) and the prebuilt
+          # .node addons — emitting hundreds of harmless "patchelf: wrong ELF
+          # type" lines per build. The real binary is node (external, already
+          # RPATH-fixed in its own derivation) and the .node addons are
+          # self-contained prebuilts loaded via dlopen, so Nix's fixup pass is
+          # neither needed nor wanted here. Skip it entirely.
+          dontFixup = true;
+
           buildPhase = pnpmInstall + ''
             echo "=== Building native voice deps ==="
             # pnpm rebuild aborts on the first failing package and runs scripts
             # from the wrong cwd — build each native dep explicitly with its own
             # install script. Each failure is tolerated (|| true); the packages
-            # that matter (opus) are verified at runtime.
+            # @discordjs/opus ships prebuilt binaries for Node 22 (ABI node-v127,
-            for pkg in \
+            # linux-x64-glibc-2.35) — node-pre-gyp downloads the prebuilt .node
-              node_modules/.pnpm/@discordjs+opus@*/node_modules/@discordjs/opus
+            # instead of compiling C++ from source. With build_from_source unset
-            do
+            # (above), `pnpm rebuild` runs the package's own install script which
-              if [ -d "$pkg" ]; then
+            # fetches the matching prebuilt; it only falls back to a source build
-                echo "--- native build: $pkg ---"
+            # if the download fails. This keeps voice working without a per-build
-                (cd "$pkg" && npm run install 2>&1 || true)
+            # native compile.
-              fi
+            echo "=== Rebuilding @discordjs/opus (prebuilt download) ==="
-            done
+            pnpm rebuild @discordjs/opus 2>&1 || true
-            echo "=== Building libdatachannel-min N-API binding ==="
-            # The GoLive screen-share stack uses a minimal N-API binding
-            # (native/libdatachannel-min) over nixpkgs libdatachannel.
-            (
-              cd native/libdatachannel-min
-              # binding.gyp resolves include/lib from env (LDC_INCLUDE = .dev
-              # include root, LDC_LIB = lib output dir, NAPI_INCLUDE =
-              # node-addon-api include root).
-              NAPI_INCLUDE=$(find ../../node_modules/.pnpm -maxdepth 3 \
-                -type d -path "*node_modules/node-addon-api" | head -1)
-              echo "NAPI_INCLUDE=$NAPI_INCLUDE"
-              LDC_INCLUDE=${libdatachannel.dev} LDC_LIB=${libdatachannel.out}/lib/libdatachannel.so.0.24.1 \
-                NAPI_INCLUDE=$NAPI_INCLUDE \
-                npx node-gyp rebuild 2>&1 || true
-              ls -la build/Release/datachannel_min.node 2>/dev/null \
-                && echo "libdatachannel-min binding OK: $(stat -c%s build/Release/datachannel_min.node) bytes" \
-                || echo "WARN: libdatachannel-min binding build FAILED (screen share disabled)"
-            )
             echo "=== Compiling TypeScript ===="
             npx tsc 2>&1
-            echo "=== Fixing @/ path aliases to relative paths ==="
+            echo "=== Fixing @/ path aliases + extensionless relative imports for node ESM ==="
-            node -e "
+            node scripts/fix-imports.mjs
-              const fs = require('fs');
-              const path = require('path');
-              let count = 0;
-              function walk(dir) {
-                if (!fs.existsSync(dir)) return;
-                for (const e of fs.readdirSync(dir, {withFileTypes: true})) {
-                  const p = path.join(dir, e.name);
-                  if (e.isDirectory()) walk(p);
-                  else if (e.name.endsWith('.js')) {
-                    const c = fs.readFileSync(p, 'utf8');
-                    const pat = /from\s+['\"]@\/([^'\"]+)['\"]/g;
-                    const n = c.replace(pat, (m, p1) => {
-                      const target = path.join('dist', p1) + '.js';
-                      const rel = path.relative(path.dirname(p), target);
-                      return 'from \"' + (rel.startsWith('.') ? rel : './' + rel) + '\"';
-                    });
-                    if (n !== c) { fs.writeFileSync(p, n); count++; }
-                  }
-                }
-              }
-              walk('dist');
-              console.log('Fixed ' + count + ' files');
-            "
             echo "=== Build complete ==="
           '' + pruneProd;
 
@@ -247,22 +188,6 @@ WRAPPER
             mkdir -p $out/lib/gmw-discord-gateway
             cp -r dist node_modules package.json tsconfig.json $out/lib/gmw-discord-gateway/
 
-            # GoLive native binding — loadNative resolves it relative to
-            # dist/goLive/native.js, i.e. <root>/native/libdatachannel-min/
-            # build/Release/datachannel_min.node; libdatachannel .so must sit
-            # next to it and be on LD_LIBRARY_PATH at runtime.
-            mkdir -p $out/lib/gmw-discord-gateway/native/libdatachannel-min/build/Release
-            cp native/libdatachannel-min/build/Release/datachannel_min.node \
-              $out/lib/gmw-discord-gateway/native/libdatachannel-min/build/Release/ 2>/dev/null || true
-            mkdir -p $out/lib/gmw-discord-gateway/native/libdatachannel-min/build/ldc
-            cp -rL native/libdatachannel-min/build/ldc/libdatachannel.so* \
-              $out/lib/gmw-discord-gateway/native/libdatachannel-min/build/ldc/ 2>/dev/null || true
-            # If the binding failed to build, screen share is simply disabled —
-            # the gateway itself must still start.
-            if [ ! -f $out/lib/gmw-discord-gateway/native/libdatachannel-min/build/Release/datachannel_min.node ]; then
-              echo "WARN: datachannel_min.node missing — GoLive screen share disabled in this build"
-            fi
-
             # Also include drizzle migrations if they exist
             cp -r drizzle $out/lib/gmw-discord-gateway/ 2>/dev/null || true
 
@@ -271,7 +196,6 @@ WRAPPER
 #!${pkgs.runtimeShell}
 cd $out/lib/gmw-discord-gateway
 export PATH=${pkgs.ffmpeg-headless}/bin:${pkgs.yt-dlp}/bin:\$PATH
-export LD_LIBRARY_PATH=${libdatachannel.out}/lib:\$LD_LIBRARY_PATH
 exec ${nodejs}/bin/node dist/index.js
 WRAPPER
             chmod +x $out/bin/gmw-discord-gateway

infra/nix/nginx.conf.template

@@ -61,6 +61,25 @@ http {
       proxy_send_timeout 86400s;
     }
 
+    # ── Backend oRPC (structured data RPCs over WebSocket + HTTP POST)
+    #     Browser reaches this via partysocket (wss://…/trpc); SSR/RSC uses
+    #     the fetch RPCLink (POST /trpc). Same path, same backend handler:
+    #     oRPC's RPCHandler (HTTP) + ORPCWebSocketServer (WS) on :4001. ──
+    location ^~ /trpc {
+      proxy_pass http://gmw_backend$uri$is_args$args;
+      proxy_http_version 1.1;
+      # Upgrade headers required for the WebSocket transport; harmless for POST.
+      proxy_set_header Upgrade              $http_upgrade;
+      proxy_set_header Connection           $connection_upgrade;
+      proxy_set_header Host                 $host;
+      proxy_set_header X-Real-IP            $remote_addr;
+      proxy_set_header X-Forwarded-For      $proxy_add_x_forwarded_for;
+      proxy_set_header X-Forwarded-Proto    $scheme;
+      proxy_buffering off;
+      proxy_read_timeout 86400s;
+      proxy_send_timeout 86400s;
+    }
+
     # ── Next.js build assets — immutable, edge/shareable ───────────
     location ^~ /_next/static/ {
       proxy_pass http://gmw_next$uri$is_args$args;

scripts/add-ai-analysis-duration.sql

@@ -0,0 +1,10 @@
+-- Migration: add ai_analysis_duration_ms to messages
+-- Tracks how long the AI moderation LLM call took, per message (ms).
+-- Idempotent: safe to re-run.
+--
+-- Run against the production GMW database, e.g.:
+--   PGPASSWORD=*** psql -h 100.121.180.82 -p 6432 -U asephs -d dcbot \
+--     -f scripts/add-ai-analysis-duration.sql
+
+ALTER TABLE "messages"
+  ADD COLUMN IF NOT EXISTS "ai_analysis_duration_ms" BIGINT;

scripts/drop-materi-documents.sql

@@ -0,0 +1,14 @@
+-- Migration: Drop materi_documents table (feature removed)
+-- Run: PGPASSWORD=<pw> psql -h <host> -U <user> -d <db> -f scripts/drop-materi-documents.sql
+-- Reverses scripts/add-materi-documents.sql which was deleted with the feature.
+
+BEGIN;
+
+DROP INDEX IF EXISTS idx_materi_search;
+DROP INDEX IF EXISTS idx_materi_guild;
+DROP INDEX IF EXISTS idx_materi_owner;
+DROP INDEX IF EXISTS idx_materi_category;
+
+DROP TABLE IF EXISTS public.materi_documents;
+
+COMMIT;

services/backend/package.json

@@ -15,6 +15,7 @@
   },
   "dependencies": {
     "@discordjs/voice": "^0.19.2",
+    "@orpc/server": "1.15.0",
     "axios": "^1.16.1",
     "dotenv": "^17.4.2",
     "drizzle-orm": "^0.45.2",
@@ -31,10 +32,10 @@
     "@biomejs/biome": "latest",
     "@types/express": "^5.0.6",
     "@types/node": "^25.9.0",
+    "@types/pg": "^8.20.0",
     "@types/ws": "^8.18.1",
     "tsx": "^4.22.2",
     "typescript": "^5.9.3",
-    "@types/pg": "^8.20.0",
     "vitest": "latest"
   }
 }

services/backend/scripts/fix-imports.mjs

@@ -0,0 +1,58 @@
+// Rewrite import specifiers in the compiled dist/ so the output runs under
+// plain `node dist/index.js` (native ESM, no bundler / no tsx).
+//
+// Background: tsconfig uses moduleResolution:"bundler", so `tsc` emits BARE
+// relative specifiers WITHOUT extensions (e.g. `import "./router"`) and leaves
+// the `@/*` path-alias imports untouched. Node's native ESM resolver rejects
+// extensionless relative specifiers and knows nothing about the `@/` alias, so
+// the emitted dist/ crashes at startup (`ERR_MODULE_NOT_FOUND`). This script
+// fixes both:
+//   1. `@/foo`            -> relative path to dist/foo.js
+//   2. `./foo` / `../foo` -> `./foo.js` / `../foo.js` (append .js)
+// Already-extensioned relative imports (.js/.json/.node/.mjs/.cjs) and bare
+// package specifiers are left untouched (idempotent).
+import { readFileSync, writeFileSync, existsSync, readdirSync } from "node:fs";
+import { join, relative, dirname } from "node:path";
+
+let count = 0;
+function walk(dir) {
+  if (!existsSync(dir)) return;
+  for (const e of readdirSync(dir, { withFileTypes: true })) {
+    const p = join(dir, e.name);
+    if (e.isDirectory()) walk(p);
+    else if (e.name.endsWith(".js")) {
+      const c = readFileSync(p, "utf8");
+      const pat = /from\s+['"]([^'"]+)['"]/g;
+      const n = c.replace(pat, (m, spec) => {
+        if (spec.startsWith("@/")) {
+          // Source may already carry an extension (e.g. "@/shared/config/index.js");
+          // only append ".js" when the specifier has none — otherwise we'd
+          // produce "index.js.js".
+          const core = spec.slice(2);
+          let target;
+          if (/\.(js|json|node|mjs|cjs)$/.test(core)) {
+            target = join("dist", core);
+          } else {
+            target = join("dist", core) + ".js";
+          }
+          let rel = relative(dirname(p), target);
+          if (!rel.startsWith(".")) rel = "./" + rel;
+          return `from "${rel}"`;
+        }
+        if (
+          (spec.startsWith("./") || spec.startsWith("../")) &&
+          !/\.(js|json|node|mjs|cjs)$/.test(spec)
+        ) {
+          return `from "${spec}.js"`;
+        }
+        return m;
+      });
+      if (n !== c) {
+        writeFileSync(p, n);
+        count++;
+      }
+    }
+  }
+}
+walk("dist");
+console.log(`Fixed ${count} import specifiers in dist/`);

services/backend/src/http/app.ts

@@ -1,3 +1,5 @@
+import { onError } from "@orpc/server";
+import { RPCHandler } from "@orpc/server/node";
 import express, {
   type Express,
   type NextFunction,
@@ -6,20 +8,17 @@ import express, {
 } from "express";
 import helmet from "helmet";
 import { createChildLogger } from "@/shared/logger/index";
-import { createAnalysisRouter } from "../modules/analysis/index.js";
-import { createChatbotRouter } from "../modules/chatbot/index.js";
-import { createConfigRouter } from "../modules/config/index.js";
-import { createDashboardRouter } from "../modules/dashboard/index.js";
 import { createHealthRouter } from "../modules/health/index.js";
-import { createMediaRouter } from "../modules/media/index.js";
+import { appRouter } from "../orpc/router";
-import { createMessagesRouter } from "../modules/messages/index.js";
-import { createModerationRouter } from "../modules/moderation/index.js";
-import { createRecordingsRouter } from "../modules/recordings/index.js";
-import { createUiStateRouter } from "../modules/ui-state/index.js";
-import { createVoiceRouter } from "../modules/voice/index.js";
 import { errorHandler } from "../shared/middlewares/index.js";
 
-// Auth removed — dashboard is public
+// Auth removed — dashboard is public.
+// All data APIs (dashboard, messages, moderation, media, voice, recordings,
+// analysis, chatbot, config, ui-state) now flow over oRPC, served on TWO
+// transports sharing the /trpc path:
+//   - WebSocket (browser live RPCs)  — see orpc/ws.ts
+//   - HTTP POST   (server-side / RSC fetch) — handled below
+// Only infra endpoints (health, prometheus metrics) remain plain HTTP.
 
 const logger = createChildLogger("http.app");
 
@@ -33,7 +32,7 @@ export function createHttpApp(): Express {
     }),
   );
 
-  // Body parsing
+  // Body parsing (still needed for any JSON POST; oRPC is WS/HTTP-based)
   app.use(express.json());
   app.use(express.urlencoded({ extended: true }));
 
@@ -59,24 +58,38 @@ export function createHttpApp(): Express {
     next();
   });
 
-  // All routes are public
+  // Infra-only HTTP endpoints
   app.use("/api", createHealthRouter());
-  app.use("/api", createConfigRouter());
+
-  app.use("/api", createDashboardRouter());
+  // oRPC over HTTP (server-side / RSC fetch). The same appRouter the browser
-  app.use("/api", createMessagesRouter());
+  // reaches over the /trpc WebSocket. oRPC's node RPCHandler writes the full
-  app.use("/api", createAnalysisRouter());
+  // response itself; if no procedure matched we fall through to the 404 below.
-  app.use("/api", createChatbotRouter());
+  const orpcHandler = new RPCHandler(appRouter, {
-  app.use("/api", createRecordingsRouter());
+    interceptors: [onError((error) => logger.error({ error }, "oRPC error"))],
-  app.use("/api", createUiStateRouter());
+  });
-  app.use("/api", createMediaRouter());
+
-  app.use("/api", createVoiceRouter());
+  app.use((req: Request, res: Response, next: NextFunction) => {
-  app.use("/api", createModerationRouter());
+    if (!req.path.startsWith("/trpc")) {
+      next();
+      return;
+    }
+    orpcHandler
+      .handle(req, res, { prefix: "/trpc", context: {} })
+      .then(({ matched }) => {
+        if (!matched) next();
+      })
+      .catch((err: unknown) => {
+        logger.error({ err }, "oRPC HTTP handler failed");
+        if (!res.headersSent) res.status(500).json({ error: "INTERNAL" });
+      });
+  });
 
   // 404 handler
   app.use((_req: Request, res: Response) => {
     res.status(404).json({
       error: "NOT_FOUND",
-      message: "Endpoint not found",
+      message:
+        "Endpoint not found — data APIs are served over /trpc (WebSocket/HTTP)",
     });
   });
 

services/backend/src/http/server.ts

@@ -1,5 +1,6 @@
 import { createServer, type Server } from "node:http";
 import { createChildLogger } from "@/shared/logger/index";
+import { createORPCWebSocketServer } from "../orpc/ws.js";
 import { config } from "../shared/config/index.js";
 import { initializeDatabase } from "../shared/database/index.js";
 import { startRedisBridge } from "../ws/redis-bridge.js";
@@ -16,8 +17,9 @@ export async function startHttpServer(): Promise<Server> {
 
   const server = createServer(app);
 
-  // Attach WebSocket server to the same HTTP server
+  // Attach WebSocket servers to the same HTTP server
-  createWebSocketServer(server);
+  createWebSocketServer(server); // /ws — voice PCM + gateway events
+  createORPCWebSocketServer(server); // /trpc — structured data RPCs
 
   // Start Redis pub/sub bridge to forward discord-gateway events to WS clients
   await startRedisBridge();

services/backend/src/modules/analysis/analysis.routes.ts

@@ -1,27 +0,0 @@
-import type { Request, Response, Router } from "express";
-import express from "express";
-import { createChildLogger } from "@/shared/logger/index";
-import { asyncHandler } from "../../shared/middlewares/index.js";
-import { analysisService } from "./analysis.service.js";
-
-const logger = createChildLogger("analysis.routes");
-
-export function createAnalysisRouter(): Router {
-  const router = express.Router();
-
-  // GET /api/analysis/search
-  router.get(
-    "/analysis/search",
-    asyncHandler(async (req: Request, res: Response) => {
-      const q = (req.query.q as string) || "";
-      const channelId = (req.query.channelId as string) || undefined;
-      const limit = Number(req.query.limit) || 20;
-
-      logger.debug({ q, channelId, limit }, "Analysis search requested");
-      const result = await analysisService.search({ q, channelId, limit });
-      res.json(result);
-    }),
-  );
-
-  return router;
-}

services/backend/src/modules/analysis/index.ts

@@ -1 +0,0 @@
-export { createAnalysisRouter } from "./analysis.routes.js";

services/backend/src/modules/chatbot/chatbot.controller.ts

@@ -1,96 +0,0 @@
-import type { Request, Response } from "express";
-import { createChildLogger } from "@/shared/logger/index";
-import { asyncHandler } from "../../shared/middlewares/index.js";
-import { chatbotService } from "./chatbot.service.js";
-
-const logger = createChildLogger("chatbot.controller");
-
-interface AuthenticatedRequest extends Request {
-  userId?: string;
-}
-
-/**
- * Resolve the actor id for a request. Frontend (no-login) sends a per-device
- * UUID via X-User-Id so chat history stays isolated per visitor; a registered
- * auth middleware userId takes precedence when present.
- */
-function resolveUserId(req: Request): string {
-  const authId = (req as AuthenticatedRequest).userId;
-  if (authId) return authId;
-  const header = (req.headers["x-user-id"] as string | undefined)?.trim();
-  return header || "anonymous";
-}
-
-export const handleChatbotChat = asyncHandler(
-  async (req: Request, res: Response) => {
-    const { message, context } = req.body as {
-      message: string;
-      context?: Record<string, unknown>;
-    };
-
-    // Validate required fields
-    if (!message || typeof message !== "string") {
-      return res.status(400).json({
-        error: "INVALID_INPUT",
-        message: "Message is required and must be a string",
-      });
-    }
-
-    // Get user ID from X-User-Id header (no-login device uuid) or auth
-    const userId = resolveUserId(req);
-
-    logger.debug(
-      { userId, messageLength: message.length, context },
-      "Received chatbot chat message",
-    );
-
-    // Process message & generate response
-    const response = await chatbotService.processMessage(
-      message,
-      context,
-      userId,
-    );
-
-    // Save conversation to database
-    await chatbotService.saveConversation({
-      userId,
-      userMessage: message,
-      botResponse: response,
-      context,
-      timestamp: new Date(),
-    });
-
-    logger.info({ userId }, "Chatbot chat processed successfully");
-
-    res.status(200).json({
-      response,
-      timestamp: new Date().toISOString(),
-    });
-  },
-);
-
-export const getChatbotHistory = asyncHandler(
-  async (req: Request, res: Response) => {
-    const userId = resolveUserId(req);
-    const limit = Math.min(parseInt(req.query.limit as string, 10) || 50, 100);
-
-    const history = await chatbotService.getChatHistory(userId, limit);
-
-    res.status(200).json({
-      history,
-      total: history.length,
-    });
-  },
-);
-
-export const clearChatbotHistory = asyncHandler(
-  async (req: Request, res: Response) => {
-    const userId = resolveUserId(req);
-
-    await chatbotService.clearChatHistory(userId);
-
-    res.status(200).json({
-      message: "Chat history cleared successfully",
-    });
-  },
-);

services/backend/src/modules/chatbot/chatbot.repository.ts

@@ -1,6 +1,6 @@
-import { and, desc, eq, type SQL, sql } from "drizzle-orm";
+import { desc, eq } from "drizzle-orm";
 import { getDatabase } from "../../shared/database/index.js";
-import { pgChatbotMessagesTable, pgMessagesTable } from "../../shared/index.js";
+import { pgChatbotMessagesTable } from "../../shared/index.js";
 import { createChildLogger } from "../../shared/logger/index.js";
 
 const logger = createChildLogger("chatbot.repository");
@@ -31,13 +31,6 @@ export interface ChatbotHistoryRow {
   created_at: string;
 }
 
-export interface ServerInsights {
-  total_messages: number;
-  active_users: number;
-  flagged: number;
-  warned: number;
-}
-
 export class ChatbotRepository {
   async saveConversation(input: SaveConversationInput): Promise<void> {
     const db = getDatabase();
@@ -83,56 +76,6 @@ export class ChatbotRepository {
       "Chat history cleared",
     );
   }
-
-  async getServerInsights(
-    guildId?: string,
-    channelId?: string,
-  ): Promise<ServerInsights> {
-    try {
-      const db = getDatabase();
-      const conditions: SQL[] = [];
-
-      if (guildId) {
-        conditions.push(eq(pgMessagesTable.guild_id, guildId));
-      }
-      if (channelId) {
-        conditions.push(eq(pgMessagesTable.channel_id, channelId));
-      }
-
-      const where = conditions.length > 0 ? and(...conditions) : undefined;
-
-      const [result] = await db
-        .select({
-          total_messages: sql<number>`COUNT(*)::int`,
-          active_users: sql<number>`COUNT(DISTINCT ${pgMessagesTable.user_id})::int`,
-          flagged: sql<number>`COUNT(*) FILTER (WHERE ${pgMessagesTable.ai_status} = 'flagged')::int`,
-          warned: sql<number>`COUNT(*) FILTER (WHERE ${pgMessagesTable.ai_status} = 'warn')::int`,
-        })
-        .from(pgMessagesTable)
-        .where(where);
-
-      const insights = result ?? {
-        total_messages: 0,
-        active_users: 0,
-        flagged: 0,
-        warned: 0,
-      };
-
-      logger.debug({ guildId, channelId, insights }, "Server insights fetched");
-      return insights;
-    } catch (error) {
-      logger.warn(
-        { error, guildId, channelId },
-        "Failed to load server insights",
-      );
-      return {
-        total_messages: 0,
-        active_users: 0,
-        flagged: 0,
-        warned: 0,
-      };
-    }
-  }
 }
 
 export const chatbotRepository = new ChatbotRepository();

services/backend/src/modules/chatbot/chatbot.routes.ts

@@ -1,18 +0,0 @@
-import express, { type Router } from "express";
-import { validateBody } from "../../shared/middlewares/index.js";
-import {
-  clearChatbotHistory,
-  getChatbotHistory,
-  handleChatbotChat,
-} from "./chatbot.controller.js";
-import { chatRequestSchema } from "./chatbot.schema.js";
-
-export function createChatbotRouter(): Router {
-  const router = express.Router();
-
-  router.post("/chat", validateBody(chatRequestSchema), handleChatbotChat);
-  router.get("/chat/history", getChatbotHistory);
-  router.delete("/chat/history", clearChatbotHistory);
-
-  return router;
-}

services/backend/src/modules/chatbot/chatbot.service.ts

@@ -6,7 +6,8 @@ import type {
   SaveConversationInput,
 } from "./chatbot.repository.js";
 import { chatbotRepository } from "./chatbot.repository.js";
-import { executeTool, tools } from "./chatbot.tools.js";
+import { tools } from "./chatbot.toolDefs.js";
+import { executeTool } from "./chatbot.tools.js";
 
 const logger = createChildLogger("chatbot.service");
 
@@ -17,22 +18,26 @@ class ChatbotService {
     userId: string,
   ): Promise<string> {
     logger.info(
-      { userId, messageLength: message.length },
+      { userId, messageLength: message.length, context },
       "processMessage called",
     );
     const recentContext = await this.getRecentConversationContext(userId);
-    const serverInsights = await chatbotRepository.getServerInsights(
+    // Scope the agent to the server/channel the user is chatting in. We no
-      context?.guildId,
+    // longer bake server stats into the prompt — the model must pull current
-      context?.channelId,
+    // data via tools (see buildSystemPrompt), so it always answers from live
-    );
+    // numbers instead of a stale snapshot.
+    const scope = {
+      guildId: context?.guildId,
+      channelId: context?.channelId,
+    };
 
-    // Build LLM messages
+    const systemPrompt = this.buildSystemPrompt(scope);
-    const systemPrompt = this.buildSystemPrompt(serverInsights);
     const conversationHistory = this.buildHistoryMessages(recentContext);
     const llmResponse = await this.callLLM(
       systemPrompt,
       conversationHistory,
       message,
+      scope,
     );
 
     return llmResponse;
@@ -66,27 +71,29 @@ class ChatbotService {
     ]);
   }
 
-  private buildSystemPrompt(insights: {
+  private buildSystemPrompt(scope: {
-    total_messages: number;
+    guildId?: string;
-    active_users: number;
+    channelId?: string;
-    flagged: number;
-    warned: number;
   }): string {
-    return `Kamu lagi ngobrol sama chatbot Discord Watcher — temen ngobrol yang tau keadaan server.
+    const scopeLine = scope.guildId
+      ? `- Scope: kamu menjawab soal server/guild id="${scope.guildId}"${scope.channelId ? `, channel id="${scope.channelId}"` : ""}.`
+      : "- Scope: tidak ada guild spesifik — jawab umum soal server ini.";
+    return `Kamu adalah chatbot Discord Watcher — temen ngobrol yang tau keadaan server, dan kamu PUNYA AKSES ke data server lewat tools.
 
-Data server saat ini:
+${scopeLine}
-- Pesan: ${insights.total_messages}
+
-- User aktif: ${insights.active_users}
+ATURAN PENTING — JANGAN PAKAI KONTEKS STATIS:
-- Flagged: ${insights.flagged}
+- Kamu TIDAK punya hafalan soal angka server (jumlah pesan, user aktif, flagged, dll). JANGAN tebak atau karang angka.
-- Warning: ${insights.warned}
+- Untuk SEMUA pertanyaan soal data server (jumlah pesan, user aktif, channel ramai, aktivitas terbaru, pesan di-flag), WAJIB panggil tool yang sesuai (get_server_stats, get_top_channels, get_recent_activity, get_top_flagged). Jawab HANYA dari hasil tool.
+- Tool otomatis di-scope ke guild/channel di atas — kalau argumen guildId/channelId kosong, biarkan kosong (sudah otomatis ter-isi). Jangan isi ID yang kamu tebak.
+- Kalau tool balas error atau kosong, bilang aja data lagi ga ketemu, jangan karang.
 
 Gaya ngobrol:
 - Santai, hangat, kayak ngobrol sama temen
 - Pake Bahasa Indonesia sehari-hari, ga perlu kaku
 - Sesekali pake emoji wajar aja, ga berlebihan
-- Kalo ditanya sesuatu yang kamu tau dari data server, jawab pake data itu
+- Kalo ditanya di luar data server dan kamu ga tau, bilang aja terus tanya balik biar ngobrolnya jalan
-- Kalo ga tau atau ga nyambung, bilang aja terus tanya balik biar ngobrolnya jalan
+- Jangan sebut "rule", "instruksi", "prompt", "tool", atau apapun soal cara kamu berpikir
-- Jangan sebut "rule", "instruksi", "prompt" atau apapun soal cara kamu berpikir
 - Biasa aja, ga usaha lucu-lucu amat — natural`;
   }
 
@@ -106,6 +113,7 @@ Gaya ngobrol:
     systemPrompt: string,
     history: Array<{ role: "user" | "assistant"; content: string }>,
     userMessage: string,
+    scope: { guildId?: string; channelId?: string },
   ): Promise<string> {
     const apiKey = config.AI_LLM_API_KEY;
     const baseUrl = config.AI_LLM_BASE_URL;
@@ -150,7 +158,13 @@ Gaya ngobrol:
             tool_choice: "auto",
             max_tokens: 600,
             temperature: 0.4,
-            stream: true,
+            // Non-streaming: request a single complete response. 9router may
+            // still emit SSE even with stream:false, so the parser below
+            // handle both raw-JSON and SSE bodies.
+            stream: false,
+            // Disable extended thinking / reasoning tokens so the bot answers
+            // directly (ignored by non-reasoning models).
+            reasoning_effort: "none",
           },
           {
             headers: {
@@ -158,14 +172,16 @@ Gaya ngobrol:
               "Content-Type": "application/json",
             },
             timeout: 45_000,
-            // 9router returns SSE even without stream:true; force stream:true
-            // in the body and read the raw SSE text.
             responseType: "text",
           },
         );
 
-        // Parse SSE `data:` lines → content + tool_calls.
+        // Parse the body into content + tool_calls. 9router may return either
-        const { content, toolCalls } = this.parseSse(response.data as string);
+        // a single JSON object (stream:false honored) or SSE text (stream
+        // implied) — parseResponse handles both.
+        const { content, toolCalls } = this.parseResponse(
+          response.data as string,
+        );
 
         logger.debug(
           {
@@ -190,9 +206,19 @@ Gaya ngobrol:
                 },
               ],
             });
+            // Auto-scope: if the model omitted guildId/channelId, fill them
+            // from the request scope so tools query the right server without
+            // the model having to guess IDs.
+            const scopedArgs = { ...tc.args };
+            if (scope.guildId && scopedArgs.guildId == null) {
+              scopedArgs.guildId = scope.guildId;
+            }
+            if (scope.channelId && scopedArgs.channelId == null) {
+              scopedArgs.channelId = scope.channelId;
+            }
             let result = "";
             try {
-              result = await executeTool(tc.name, tc.args);
+              result = await executeTool(tc.name, scopedArgs);
             } catch (e) {
               result = `Tool error: ${(e as Error).message}`;
             }
@@ -224,6 +250,60 @@ Gaya ngobrol:
     }
   }
 
+  /**
+   * Parse an LLM HTTP body into content + tool_calls. Handles both shapes
+   * 9router can return: a single JSON object (stream:false honored) or SSE
+   * text (stream implied). For SSE we delegate to parseSse.
+   */
+  private parseResponse(body: string): {
+    content: string;
+    toolCalls: Array<{
+      id: string;
+      name: string;
+      arguments: string;
+      args: Record<string, unknown>;
+    }>;
+  } {
+    const trimmed = body.trim();
+    // Non-streaming response: a single JSON object.
+    if (trimmed.startsWith("{")) {
+      try {
+        const json = JSON.parse(trimmed) as {
+          choices?: Array<{
+            message?: {
+              content?: string | null;
+              tool_calls?: Array<{
+                id?: string;
+                type?: string;
+                function?: { name?: string; arguments?: string };
+              }>;
+            };
+            delta?: unknown;
+          }>;
+        };
+        const msg = json.choices?.[0]?.message;
+        // If the router returned SSE-style shape under `choices[].delta`
+        // (rare), fall through to the SSE parser.
+        if (msg) {
+          const content = msg.content ?? "";
+          const toolCalls = (msg.tool_calls ?? []).map((tc, i) => {
+            const id = tc.id || `tool_${i}_${Date.now()}`;
+            return {
+              id,
+              name: tc.function?.name ?? "",
+              arguments: tc.function?.arguments ?? "",
+              args: this.safeJsonParse(tc.function?.arguments ?? ""),
+            };
+          });
+          return { content: content.trim(), toolCalls };
+        }
+      } catch {
+        // Not valid JSON after all — treat as SSE below.
+      }
+    }
+    return this.parseSse(body);
+  }
+
   /**
    * Parse an SSE stream body into accumulated content + any tool_calls.
    * 9router (and most OpenAI-compatible routers) emit `data: {json}` lines

services/backend/src/modules/chatbot/chatbot.toolDefs.ts

@@ -0,0 +1,270 @@
+/**
+ * Static tool *definitions* for the chatbot LLM (OpenAI function-calling
+ * format). Kept separate from the executor (chatbot.tools.ts) so the schema
+ * the model depends on can be imported without pulling in the database /
+ * config layer.
+ *
+ * The chatbot is a server-watcher agent: it can answer about ANY server
+ * situation — activity, moderation queue, specific users, channels, voice
+ * recordings, AI correction history, and trends over time — by calling these
+ * tools, which the executor implements against real tables.
+ */
+
+export interface ToolDef {
+  type: "function";
+  function: {
+    name: string;
+    description: string;
+    parameters: {
+      type: "object";
+      properties: Record<string, unknown>;
+      required?: string[];
+    };
+  };
+}
+
+export const tools: ToolDef[] = [
+  {
+    type: "function",
+    function: {
+      name: "get_server_stats",
+      description:
+        "Ambil statistik ringkas server/guild: total pesan, user aktif, jumlah pesan flagged, warn, dan clean. Panggil untuk jawab pertanyaan umum soal kondisi server. guildId/channelId otomatis ter-isi dari scope; kosongkan untuk semua data.",
+      parameters: {
+        type: "object",
+        properties: {
+          guildId: { type: "string", description: "ID server (opsional)." },
+          channelId: { type: "string", description: "ID channel (opsional)." },
+        },
+      },
+    },
+  },
+  {
+    type: "function",
+    function: {
+      name: "get_top_channels",
+      description:
+        "Ambil daftar channel paling aktif (jumlah pesan terbanyak). Panggil untuk 'channel mana paling ramai' atau aktivitas per-channel.",
+      parameters: {
+        type: "object",
+        properties: {
+          guildId: { type: "string", description: "ID server (opsional)." },
+          limit: {
+            type: "number",
+            description: "Jumlah channel teratas (default 5, max 10).",
+          },
+        },
+      },
+    },
+  },
+  {
+    type: "function",
+    function: {
+      name: "get_recent_activity",
+      description:
+        "Ambil pesan terbaru di server: siapa, di channel mana, jam berapa, isinya. Panggil untuk 'lagi ngapain' / aktivitas terbaru.",
+      parameters: {
+        type: "object",
+        properties: {
+          guildId: { type: "string", description: "ID server (opsional)." },
+          channelId: { type: "string", description: "ID channel (opsional)." },
+          limit: {
+            type: "number",
+            description: "Jumlah pesan terakhir (default 5, max 20).",
+          },
+        },
+      },
+    },
+  },
+  {
+    type: "function",
+    function: {
+      name: "get_top_flagged",
+      description:
+        "Ambil pesan dengan ai_status flagged (beserta alasan, severity, analysis). Panggil untuk bahas pesan bermasalah / kerjaan moderator.",
+      parameters: {
+        type: "object",
+        properties: {
+          guildId: { type: "string", description: "ID server (opsional)." },
+          channelId: { type: "string", description: "ID channel (opsional)." },
+          limit: { type: "number", description: "Jumlah pesan (default 5)." },
+        },
+      },
+    },
+  },
+  {
+    type: "function",
+    function: {
+      name: "search_messages",
+      description:
+        "Cari pesan berdasarkan kata kunci di isi pesan (case-insensitive, LIKE). Untuk 'ada yang bahas X gak?' / temukan topik tertentu. Hindari kata terlalu umum.",
+      parameters: {
+        type: "object",
+        properties: {
+          query: {
+            type: "string",
+            description: "Kata kunci pencarian (wajib).",
+          },
+          guildId: { type: "string", description: "ID server (opsional)." },
+          channelId: { type: "string", description: "ID channel (opsional)." },
+          limit: { type: "number", description: "Jumlah hasil (default 5)." },
+        },
+        required: ["query"],
+      },
+    },
+  },
+  {
+    type: "function",
+    function: {
+      name: "get_user_messages",
+      description:
+        "Ambil pesan terbaru dari satu user tertentu (user_id), opsional di-scope ke guild/channel. Untuk 'chat si A gimana akhir-akhir ini?' — butuh user_id.",
+      parameters: {
+        type: "object",
+        properties: {
+          userId: { type: "string", description: "ID user (wajib)." },
+          guildId: { type: "string", description: "ID server (opsional)." },
+          channelId: { type: "string", description: "ID channel (opsional)." },
+          limit: { type: "number", description: "Jumlah pesan (default 10)." },
+        },
+        required: ["userId"],
+      },
+    },
+  },
+  {
+    type: "function",
+    function: {
+      name: "get_user_profile",
+      description:
+        "Ambil ringkasan profil AI dari seorang user (pola perilaku, gaya bicara) dari tabel user_profiles. Untuk 'siapa si A?' / konteks perilaku. Butuh user_id.",
+      parameters: {
+        type: "object",
+        properties: {
+          userId: { type: "string", description: "ID user (wajib)." },
+          guildId: { type: "string", description: "ID server (opsional)." },
+        },
+        required: ["userId"],
+      },
+    },
+  },
+  {
+    type: "function",
+    function: {
+      name: "get_user_reputation",
+      description:
+        "Ambil skor trust, jumlah infraction, dan streak pesan bersih seorang user dari user_reputations. Untuk 'berapa trust score si A?' / riwayat pelanggaran. Butuh user_id.",
+      parameters: {
+        type: "object",
+        properties: {
+          userId: { type: "string", description: "ID user (wajib)." },
+          guildId: { type: "string", description: "ID server (opsional)." },
+        },
+        required: ["userId"],
+      },
+    },
+  },
+  {
+    type: "function",
+    function: {
+      name: "get_channel_culture",
+      description:
+        "Ambil ringkasan norma/slang channel dari tabel channel_cultures (AI-generated). Untuk 'norma channel ini gimana?' / konteks sebelum nge-flag. Butuh channel_id.",
+      parameters: {
+        type: "object",
+        properties: {
+          channelId: { type: "string", description: "ID channel (wajib)." },
+        },
+        required: ["channelId"],
+      },
+    },
+  },
+  {
+    type: "function",
+    function: {
+      name: "get_message_detail",
+      description:
+        "Ambil 1 pesan lengkap beserta hasil analisis AI-nya (status, flags, score, severity, kategori, analysis, recommended action). Untuk jelasin keputusan moderasi pada pesan tertentu. Butuh message_id.",
+      parameters: {
+        type: "object",
+        properties: {
+          messageId: { type: "string", description: "ID pesan (wajib)." },
+        },
+        required: ["messageId"],
+      },
+    },
+  },
+  {
+    type: "function",
+    function: {
+      name: "get_message_reviews",
+      description:
+        "Ambil antrean review moderasi manual (message_reviews) berdasarkan status: pending/approved/rejected/escalated. Untuk 'ada review moderasi pending?' / cek kerjaan human moderator. guildId otomatis ter-isi.",
+      parameters: {
+        type: "object",
+        properties: {
+          guildId: { type: "string", description: "ID server (opsional)." },
+          status: {
+            type: "string",
+            description:
+              "Status review: pending / approved / rejected / escalated (opsional, default semua).",
+          },
+          limit: { type: "number", description: "Jumlah (default 10)." },
+        },
+      },
+    },
+  },
+  {
+    type: "function",
+    function: {
+      name: "get_voice_recordings",
+      description:
+        "Ambil rekaman suara terbaru (voice_recordings): user, channel, transkripsi, status upload. Untuk 'ada rekaman suara terbaru?' / cek transkripsi. Bisa di-scope ke user_id atau channel_id.",
+      parameters: {
+        type: "object",
+        properties: {
+          userId: { type: "string", description: "Filter user (opsional)." },
+          channelId: {
+            type: "string",
+            description: "Filter channel (opsional).",
+          },
+          guildId: { type: "string", description: "ID server (opsional)." },
+          limit: { type: "number", description: "Jumlah (default 10)." },
+        },
+      },
+    },
+  },
+  {
+    type: "function",
+    function: {
+      name: "get_moderation_timeline",
+      description:
+        "Ambil tren harian: per hari, jumlah total pesan vs flagged vs warn vs clean. Untuk 'minggu ini pelanggaran naik?' / lihat tren moderasi. guildId otomatis ter-isi.",
+      parameters: {
+        type: "object",
+        properties: {
+          guildId: { type: "string", description: "ID server (opsional)." },
+          channelId: { type: "string", description: "ID channel (opsional)." },
+          days: {
+            type: "number",
+            description: "Jumlah hari ke belakang (default 14, max 60).",
+          },
+        },
+      },
+    },
+  },
+  {
+    type: "function",
+    function: {
+      name: "get_corrections",
+      description:
+        "Ambil riwayat koreksi false-positive AI (corrected_moderations): pesan yang awalnya di-flag tapi dikoreksi manusia, beserta alasannya. Untuk 'AI pernah salah nge-flag apa aja?' / audit akurasi moderasi.",
+      parameters: {
+        type: "object",
+        properties: {
+          guildId: { type: "string", description: "ID server (opsional)." },
+          limit: { type: "number", description: "Jumlah (default 10)." },
+        },
+      },
+    },
+  },
+];

services/backend/src/modules/chatbot/chatbot.tools.ts

@@ -1,116 +1,25 @@
-import { sql } from "drizzle-orm";
+import { and, desc, eq, like, sql } from "drizzle-orm";
 import { getDatabase } from "../../shared/database/index.js";
+import {
+  pgChannelCulturesTable,
+  pgCorrectedModerationsTable,
+  pgMessageReviewsTable,
+  pgMessagesTable,
+  pgUserProfilesTable,
+  pgUserReputationsTable,
+  pgVoiceRecordingsTable,
+} from "../../shared/index.js";
 
 /**
- * Tools the chatbot LLM can call. Definitions describe the schema to the
+ * Executor for the chatbot's server-watcher tools. The tool *definitions*
- * model; the executor implements each one against the real database.
+ * live in chatbot.toolDefs.ts (no DB import); this file implements each one
- * This turns the chatbot from "blind stats guesser" into an agent that
+ * against the real database.
- * pulls real, current server data on demand.
+ *
+ * All queries use parameterized drizzle operators (eq/like/and) — never string
+ * interpolation into raw SQL — so model-supplied arguments cannot inject SQL.
  */
 
 export type ToolResult = string;
-
-/** JSON schema for a tool definition (OpenAI function-calling format). */
-export interface ToolDef {
-  type: "function";
-  function: {
-    name: string;
-    description: string;
-    parameters: {
-      type: "object";
-      properties: Record<string, unknown>;
-      required?: string[];
-    };
-  };
-}
-
-export const tools: ToolDef[] = [
-  {
-    type: "function",
-    function: {
-      name: "get_server_stats",
-      description:
-        "Ambil statistik ringkas server/guild saat ini: total pesan, user aktif, jumlah pesan flagged, dan jumlah warning. Panggil ini untuk menjawab pertanyaan umum tentang kondisi server. Opsional fill guild_id untuk scope ke guild tertentu, channel_id untuk scope ke channel.",
-      parameters: {
-        type: "object",
-        properties: {
-          guildId: {
-            type: "string",
-            description: "ID guild/server (opsional). Kosongkan = semua data.",
-          },
-          channelId: {
-            type: "string",
-            description: "ID channel (opsional).",
-          },
-        },
-      },
-    },
-  },
-  {
-    type: "function",
-    function: {
-      name: "get_top_channels",
-      description:
-        "Ambil daftar channel paling aktif (jumlah pesan terbanyak) di server. Panggil buat jawab 'channel mana paling ramai' atau aktivitas per-channel.",
-      parameters: {
-        type: "object",
-        properties: {
-          guildId: {
-            type: "string",
-            description: "ID server (opsional).",
-          },
-          limit: {
-            type: "number",
-            description: "Jumlah channel teratas (default 5, max 10).",
-          },
-        },
-      },
-    },
-  },
-  {
-    type: "function",
-    function: {
-      name: "get_recent_activity",
-      description:
-        "Ambil aktivitas/pesan terbaru di server: siapa yang baru ngomong, di channel mana, jam berapa. Panggil buat jawaban soal 'lagi ngapain' / aktivitas terbaru di server.",
-      parameters: {
-        type: "object",
-        properties: {
-          guildId: {
-            type: "string",
-            description: "ID server (opsional).",
-          },
-          limit: {
-            type: "number",
-            description: "Jumlah pesan terakhir (default 5).",
-          },
-        },
-      },
-    },
-  },
-  {
-    type: "function",
-    function: {
-      name: "get_top_flagged",
-      description:
-        "Ambil pesan yang paling sering di-flag atau kena warning. Panggil buat jawab soal pesan bermasalah / moderator.",
-      parameters: {
-        type: "object",
-        properties: {
-          guildId: {
-            type: "string",
-            description: "ID server (opsional).",
-          },
-          limit: {
-            type: "number",
-            description: "Jumlah pesan (default 5).",
-          },
-        },
-      },
-    },
-  },
-];
-
 /** Executes a tool call against the real DB and returns a readable result. */
 export async function executeTool(
   name: string,
@@ -122,9 +31,11 @@ export async function executeTool(
     typeof args.channelId === "string" && args.channelId
       ? args.channelId
       : undefined;
+  const userId =
+    typeof args.userId === "string" && args.userId ? args.userId : undefined;
   const limitRaw =
     typeof args.limit === "number" ? args.limit : Number(args.limit) || 5;
-  const limit = Math.min(Math.max(1, Math.round(limitRaw)), 10);
+  const limit = Math.min(Math.max(1, Math.round(limitRaw)), 20);
 
   try {
     switch (name) {
@@ -133,9 +44,48 @@ export async function executeTool(
       case "get_top_channels":
         return await topChannels(guildId, limit);
       case "get_recent_activity":
-        return await recentActivity(guildId, limit);
+        return await recentActivity(guildId, channelId, limit);
       case "get_top_flagged":
-        return await topFlagged(guildId, limit);
+        return await topFlagged(guildId, channelId, limit);
+      case "search_messages":
+        return await searchMessages(
+          String(args.query ?? ""),
+          guildId,
+          channelId,
+          limit,
+        );
+      case "get_user_messages":
+        return await userMessages(userId, guildId, channelId, limit);
+      case "get_user_profile":
+        return await userProfile(userId, guildId);
+      case "get_user_reputation":
+        return await userReputation(userId, guildId);
+      case "get_channel_culture":
+        return await channelCulture(
+          typeof args.channelId === "string" ? args.channelId : undefined,
+        );
+      case "get_message_detail":
+        return await messageDetail(
+          typeof args.messageId === "string" ? args.messageId : undefined,
+        );
+      case "get_message_reviews":
+        return await messageReviews(
+          guildId,
+          typeof args.status === "string" ? args.status : undefined,
+          limit,
+        );
+      case "get_voice_recordings":
+        return await voiceRecordings(userId, channelId, guildId, limit);
+      case "get_moderation_timeline":
+        return await moderationTimeline(
+          guildId,
+          channelId,
+          typeof args.days === "number"
+            ? Math.min(Math.max(1, args.days), 60)
+            : 14,
+        );
+      case "get_corrections":
+        return await corrections(guildId, limit);
       default:
         return `Unknown tool: ${name}`;
     }
@@ -145,6 +95,23 @@ export async function executeTool(
   }
 }
 
+// ── Query helpers ──────────────────────────────────────────
+
+function scopeMessages(
+  guildId?: string,
+  channelId?: string,
+): ReturnType<typeof and> | undefined {
+  const conds = [];
+  if (guildId) conds.push(eq(pgMessagesTable.guild_id, guildId));
+  if (channelId) conds.push(eq(pgMessagesTable.channel_id, channelId));
+  return conds.length ? and(...conds) : undefined;
+}
+
+/** Escape LIKE wildcards so user input can't break the pattern. */
+function likePattern(q: string): string {
+  return q.replace(/[\\%_]/g, (c) => `\\${c}`);
+}
+
 // ── Tool executors ──────────────────────────────────────────
 
 async function serverStats(
@@ -152,81 +119,330 @@ async function serverStats(
   channelId?: string,
 ): Promise<string> {
   const db = getDatabase();
-  const conditions: string[] = [];
+  const [result] = await db
-  if (guildId) conditions.push(`guild_id = '${guildId}'`);
+    .select({
-  if (channelId) conditions.push(`channel_id = '${channelId}'`);
+      total_messages: sql<number>`COUNT(*)::int`,
-  const cond = conditions.length ? `WHERE ${conditions.join(" AND ")}` : "";
+      active_users: sql<number>`COUNT(DISTINCT ${pgMessagesTable.user_id})::int`,
+      flagged: sql<number>`COUNT(*) FILTER (WHERE ${pgMessagesTable.ai_status} = 'flagged')::int`,
+      warned: sql<number>`COUNT(*) FILTER (WHERE ${pgMessagesTable.ai_status} = 'warn')::int`,
+      clean: sql<number>`COUNT(*) FILTER (WHERE ${pgMessagesTable.ai_status} = 'clean')::int`,
+    })
+    .from(pgMessagesTable)
+    .where(scopeMessages(guildId, channelId));
 
-  const result = await db.execute(
+  const r = result ?? {
-    sql.raw(
+    total_messages: 0,
-      `SELECT COUNT(*)::int AS total_messages,
+    active_users: 0,
-              COUNT(DISTINCT user_id)::int AS active_users,
+    flagged: 0,
-              COUNT(*) FILTER (WHERE ai_status = 'flagged')::int AS flagged,
+    warned: 0,
-              COUNT(*) FILTER (WHERE ai_status = 'warn')::int AS warned
+    clean: 0,
-       FROM messages ${cond}`,
+  };
-    ),
+  return JSON.stringify(r);
-  );
-  const rows =
-    (result as unknown as { rows: Record<string, unknown>[] }).rows ?? [];
-  const r = rows[0] ?? {};
-  return JSON.stringify({
-    total_messages: r.total_messages ?? 0,
-    active_users: r.active_users ?? 0,
-    flagged: r.flagged ?? 0,
-    warned: r.warned ?? 0,
-  });
 }
 
 async function topChannels(guildId?: string, limit = 5): Promise<string> {
   const db = getDatabase();
-  const conditions: string[] = [];
+  const rows = await db
-  if (guildId) conditions.push(`guild_id = '${guildId}'`);
+    .select({
-  const cond = conditions.length ? `WHERE ${conditions.join(" AND ")}` : "";
+      channel_id: pgMessagesTable.channel_id,
-
+      count: sql<number>`COUNT(*)::int`,
-  const result = await db.execute(
+    })
-    sql.raw(
+    .from(pgMessagesTable)
-      `SELECT channel_id,
+    .where(scopeMessages(guildId))
-              COUNT(*)::int AS count
+    .groupBy(pgMessagesTable.channel_id)
-       FROM messages ${cond}
+    .orderBy(desc(sql`COUNT(*)`))
-       GROUP BY channel_id
+    .limit(limit);
-       ORDER BY count DESC
+  return JSON.stringify(rows);
-       LIMIT ${limit}`,
-    ),
-  );
-  const rows = (result as unknown as { rows: unknown[] }).rows ?? [];
-  return JSON.stringify(rows.slice(0, limit));
 }
 
-async function recentActivity(guildId?: string, limit = 5): Promise<string> {
+async function recentActivity(
+  guildId?: string,
+  channelId?: string,
+  limit = 5,
+): Promise<string> {
   const db = getDatabase();
-  const conditions: string[] = [];
+  const rows = await db
-  if (guildId) conditions.push(`guild_id = '${guildId}'`);
+    .select({
-  const cond = conditions.length ? `WHERE ${conditions.join(" AND ")}` : "";
+      id: pgMessagesTable.id,
-
+      username: pgMessagesTable.username,
-  const result = await db.execute(
+      user_id: pgMessagesTable.user_id,
-    sql.raw(
+      channel_id: pgMessagesTable.channel_id,
-      `SELECT username, content, channel_id, created_at
+      content: pgMessagesTable.content,
-       FROM messages ${cond}
+      created_at: pgMessagesTable.created_at,
-       ORDER BY created_at DESC
+      ai_status: pgMessagesTable.ai_status,
-       LIMIT ${limit}`,
+    })
-    ),
+    .from(pgMessagesTable)
-  );
+    .where(scopeMessages(guildId, channelId))
-  return JSON.stringify((result as unknown as { rows: unknown[] }).rows ?? []);
+    .orderBy(desc(pgMessagesTable.created_at))
+    .limit(limit);
+  return JSON.stringify(rows);
 }
 
-async function topFlagged(guildId?: string, limit = 5): Promise<string> {
+async function topFlagged(
+  guildId?: string,
+  channelId?: string,
+  limit = 5,
+): Promise<string> {
   const db = getDatabase();
-  const conditions = ["ai_status IN ('flagged', 'warn')"];
+  const rows = await db
-  if (guildId) conditions.push(`guild_id = '${guildId}'`);
+    .select({
-  const cond = `WHERE ${conditions.join(" AND ")}`;
+      id: pgMessagesTable.id,
-
+      username: pgMessagesTable.username,
-  const result = await db.execute(
+      channel_id: pgMessagesTable.channel_id,
-    sql.raw(
+      content: pgMessagesTable.content,
-      `SELECT username, content, channel_id, ai_status, created_at
+      ai_status: pgMessagesTable.ai_status,
-       FROM messages ${cond}
+      ai_severity: pgMessagesTable.ai_severity,
-       ORDER BY created_at DESC
+      ai_moderation_flags: pgMessagesTable.ai_moderation_flags,
-       LIMIT ${limit}`,
+      ai_analysis: pgMessagesTable.ai_analysis,
+      created_at: pgMessagesTable.created_at,
+    })
+    .from(pgMessagesTable)
+    .where(
+      and(
+        scopeMessages(guildId, channelId),
+        eq(pgMessagesTable.ai_status, "flagged"),
       ),
-  );
+    )
-  return JSON.stringify((result as unknown as { rows: unknown[] }).rows ?? []);
+    .orderBy(desc(pgMessagesTable.created_at))
+    .limit(limit);
+  return JSON.stringify(rows);
+}
+
+async function searchMessages(
+  query: string,
+  guildId?: string,
+  channelId?: string,
+  limit = 5,
+): Promise<string> {
+  const db = getDatabase();
+  if (!query.trim()) return JSON.stringify({ error: "query kosong" });
+  const rows = await db
+    .select({
+      id: pgMessagesTable.id,
+      username: pgMessagesTable.username,
+      channel_id: pgMessagesTable.channel_id,
+      content: pgMessagesTable.content,
+      created_at: pgMessagesTable.created_at,
+      ai_status: pgMessagesTable.ai_status,
+    })
+    .from(pgMessagesTable)
+    .where(
+      and(
+        scopeMessages(guildId, channelId),
+        like(pgMessagesTable.content, `%${likePattern(query)}%`),
+      ),
+    )
+    .orderBy(desc(pgMessagesTable.created_at))
+    .limit(limit);
+  return JSON.stringify(rows);
+}
+
+async function userMessages(
+  userId?: string,
+  guildId?: string,
+  channelId?: string,
+  limit = 10,
+): Promise<string> {
+  const db = getDatabase();
+  if (!userId) return JSON.stringify({ error: "userId wajib" });
+  const conds = [eq(pgMessagesTable.user_id, userId)];
+  if (guildId) conds.push(eq(pgMessagesTable.guild_id, guildId));
+  if (channelId) conds.push(eq(pgMessagesTable.channel_id, channelId));
+  const rows = await db
+    .select({
+      id: pgMessagesTable.id,
+      channel_id: pgMessagesTable.channel_id,
+      content: pgMessagesTable.content,
+      created_at: pgMessagesTable.created_at,
+      ai_status: pgMessagesTable.ai_status,
+    })
+    .from(pgMessagesTable)
+    .where(and(...conds))
+    .orderBy(desc(pgMessagesTable.created_at))
+    .limit(limit);
+  return JSON.stringify(rows);
+}
+
+async function userProfile(userId?: string, guildId?: string): Promise<string> {
+  const db = getDatabase();
+  if (!userId) return JSON.stringify({ error: "userId wajib" });
+  const conds = [eq(pgUserProfilesTable.user_id, userId)];
+  if (guildId) conds.push(eq(pgUserProfilesTable.guild_id, guildId));
+  const rows = await db
+    .select({
+      user_id: pgUserProfilesTable.user_id,
+      guild_id: pgUserProfilesTable.guild_id,
+      profile_summary: pgUserProfilesTable.profile_summary,
+      last_analyzed_at: pgUserProfilesTable.last_analyzed_at,
+    })
+    .from(pgUserProfilesTable)
+    .where(and(...conds))
+    .limit(1);
+  return JSON.stringify(rows[0] ?? { error: "profil tidak ditemukan" });
+}
+
+async function userReputation(
+  userId?: string,
+  guildId?: string,
+): Promise<string> {
+  const db = getDatabase();
+  if (!userId) return JSON.stringify({ error: "userId wajib" });
+  const conds = [eq(pgUserReputationsTable.user_id, userId)];
+  if (guildId) conds.push(eq(pgUserReputationsTable.guild_id, guildId));
+  const rows = await db
+    .select({
+      user_id: pgUserReputationsTable.user_id,
+      guild_id: pgUserReputationsTable.guild_id,
+      trust_score: pgUserReputationsTable.trust_score,
+      clean_message_streak: pgUserReputationsTable.clean_message_streak,
+      total_infractions: pgUserReputationsTable.total_infractions,
+      last_infraction_at: pgUserReputationsTable.last_infraction_at,
+    })
+    .from(pgUserReputationsTable)
+    .where(and(...conds))
+    .limit(1);
+  return JSON.stringify(rows[0] ?? { error: "reputasi tidak ditemukan" });
+}
+
+async function channelCulture(channelId?: string): Promise<string> {
+  const db = getDatabase();
+  if (!channelId) return JSON.stringify({ error: "channelId wajib" });
+  const rows = await db
+    .select({
+      channel_id: pgChannelCulturesTable.channel_id,
+      culture_summary: pgChannelCulturesTable.culture_summary,
+      last_analyzed_at: pgChannelCulturesTable.last_analyzed_at,
+    })
+    .from(pgChannelCulturesTable)
+    .where(eq(pgChannelCulturesTable.channel_id, channelId))
+    .limit(1);
+  return JSON.stringify(rows[0] ?? { error: "culture tidak ditemukan" });
+}
+
+async function messageDetail(messageId?: string): Promise<string> {
+  const db = getDatabase();
+  if (!messageId) return JSON.stringify({ error: "messageId wajib" });
+  const rows = await db
+    .select({
+      id: pgMessagesTable.id,
+      guild_id: pgMessagesTable.guild_id,
+      channel_id: pgMessagesTable.channel_id,
+      user_id: pgMessagesTable.user_id,
+      username: pgMessagesTable.username,
+      content: pgMessagesTable.content,
+      created_at: pgMessagesTable.created_at,
+      ai_status: pgMessagesTable.ai_status,
+      ai_moderation_flags: pgMessagesTable.ai_moderation_flags,
+      ai_moderation_score: pgMessagesTable.ai_moderation_score,
+      ai_severity: pgMessagesTable.ai_severity,
+      ai_categories: pgMessagesTable.ai_categories,
+      ai_analysis: pgMessagesTable.ai_analysis,
+      ai_recommended_action: pgMessagesTable.ai_recommended_action,
+      ai_confidence: pgMessagesTable.ai_confidence,
+    })
+    .from(pgMessagesTable)
+    .where(eq(pgMessagesTable.id, messageId))
+    .limit(1);
+  return JSON.stringify(rows[0] ?? { error: "pesan tidak ditemukan" });
+}
+
+async function messageReviews(
+  guildId?: string,
+  status?: string,
+  limit = 10,
+): Promise<string> {
+  const db = getDatabase();
+  const conds = [];
+  if (guildId) conds.push(eq(pgMessageReviewsTable.guild_id, guildId));
+  if (status) conds.push(eq(pgMessageReviewsTable.status, status as never));
+  const rows = await db
+    .select({
+      id: pgMessageReviewsTable.id,
+      message_id: pgMessageReviewsTable.message_id,
+      reviewer_id: pgMessageReviewsTable.reviewer_id,
+      status: pgMessageReviewsTable.status,
+      notes: pgMessageReviewsTable.notes,
+      created_at: pgMessageReviewsTable.created_at,
+      reviewed_at: pgMessageReviewsTable.reviewed_at,
+    })
+    .from(pgMessageReviewsTable)
+    .where(conds.length ? and(...conds) : undefined)
+    .orderBy(desc(pgMessageReviewsTable.created_at))
+    .limit(limit);
+  return JSON.stringify(rows);
+}
+
+async function voiceRecordings(
+  userId?: string,
+  channelId?: string,
+  guildId?: string,
+  limit = 10,
+): Promise<string> {
+  const db = getDatabase();
+  const conds = [];
+  if (userId) conds.push(eq(pgVoiceRecordingsTable.user_id, userId));
+  if (channelId) conds.push(eq(pgVoiceRecordingsTable.channel_id, channelId));
+  if (guildId) conds.push(eq(pgVoiceRecordingsTable.guild_id, guildId));
+  const rows = await db
+    .select({
+      id: pgVoiceRecordingsTable.id,
+      username: pgVoiceRecordingsTable.username,
+      channel_name: pgVoiceRecordingsTable.channel_name,
+      filename: pgVoiceRecordingsTable.filename,
+      size_bytes: pgVoiceRecordingsTable.size_bytes,
+      upload_status: pgVoiceRecordingsTable.upload_status,
+      transcription: pgVoiceRecordingsTable.transcription,
+      created_at: pgVoiceRecordingsTable.created_at,
+    })
+    .from(pgVoiceRecordingsTable)
+    .where(conds.length ? and(...conds) : undefined)
+    .orderBy(desc(pgVoiceRecordingsTable.created_at))
+    .limit(limit);
+  return JSON.stringify(rows);
+}
+
+async function moderationTimeline(
+  guildId?: string,
+  channelId?: string,
+  days = 14,
+): Promise<string> {
+  const db = getDatabase();
+  const day = sql<string>`to_char(to_timestamp(${pgMessagesTable.created_at} / 1000), 'YYYY-MM-DD')`;
+  const rows = await db
+    .select({
+      day,
+      total: sql<number>`COUNT(*)::int`,
+      flagged: sql<number>`COUNT(*) FILTER (WHERE ${pgMessagesTable.ai_status} = 'flagged')::int`,
+      warned: sql<number>`COUNT(*) FILTER (WHERE ${pgMessagesTable.ai_status} = 'warn')::int`,
+      clean: sql<number>`COUNT(*) FILTER (WHERE ${pgMessagesTable.ai_status} = 'clean')::int`,
+    })
+    .from(pgMessagesTable)
+    .where(
+      and(
+        scopeMessages(guildId, channelId),
+        // only the last N days
+        sql`${pgMessagesTable.created_at} >= extract(epoch FROM now() - (${days} || ' days')::interval) * 1000`,
+      ),
+    )
+    .groupBy(day)
+    .orderBy(day);
+  return JSON.stringify(rows);
+}
+
+async function corrections(_guildId?: string, limit = 10): Promise<string> {
+  const db = getDatabase();
+  const rows = await db
+    .select({
+      id: pgCorrectedModerationsTable.id,
+      message_id: pgCorrectedModerationsTable.message_id,
+      original_flags: pgCorrectedModerationsTable.original_flags,
+      corrected_flags: pgCorrectedModerationsTable.corrected_flags,
+      correction_notes: pgCorrectedModerationsTable.correction_notes,
+      content_snippet: pgCorrectedModerationsTable.content_snippet,
+      created_at: pgCorrectedModerationsTable.created_at,
+    })
+    .from(pgCorrectedModerationsTable)
+    .orderBy(desc(pgCorrectedModerationsTable.created_at))
+    .limit(limit);
+  return JSON.stringify(rows);
 }

services/backend/src/modules/chatbot/index.ts

@@ -1 +0,0 @@
-export { createChatbotRouter } from "./chatbot.routes.js";

services/backend/src/modules/config/config.routes.ts

@@ -1,28 +0,0 @@
-import type { Router } from "express";
-import express from "express";
-import { config } from "../../shared/config/index.js";
-
-export function createConfigRouter(): Router {
-  const router = express.Router();
-
-  // GET /api/config
-  router.get("/config", (_req, res) => {
-    res.json({
-      monitorGuildId: config.MONITOR_GUILD_ID || null,
-      webserverPort: config.WEBSERVER_PORT,
-      nodeEnv: config.NODE_ENV,
-      backlogSyncHours: config.BACKLOG_SYNC_HOURS,
-      backlogSyncBatchSize: config.BACKLOG_SYNC_BATCH_SIZE,
-      retentionMessagesDays: config.RETENTION_MESSAGES_DAYS,
-      retentionAttachmentsDays: config.RETENTION_ATTACHMENTS_DAYS,
-      retentionVoiceDays: config.RETENTION_VOICE_DAYS,
-      autoDeleteFlaggedEnabled: config.AUTO_DELETE_FLAGGED_ENABLED,
-      aiAnalysisEnabled: config.AI_ANALYSIS_ENABLED,
-      voiceGuildId: config.VOICE_GUILD_ID || null,
-      voiceChannelId: config.VOICE_CHANNEL_ID || null,
-      logLevel: config.LOG_LEVEL,
-    });
-  });
-
-  return router;
-}

services/backend/src/modules/config/index.ts

@@ -1 +0,0 @@
-export { createConfigRouter } from "./config.routes.js";

services/backend/src/modules/dashboard/dashboard.repository.ts

@@ -5,7 +5,6 @@ import {
   pgChannelCulturesTable,
   pgMessagesTable,
   pgUserProfilesTable,
-  pgUserReputationsTable,
   pgVoiceRecordingsTable,
 } from "../../shared/index.js";
 import type { ListUsersQuery } from "./dashboard.service.js";
@@ -156,8 +155,7 @@ export class DashboardRepository {
         p.profile_summary,
         m.total_messages,
         m.flagged_count,
-        m.last_message_at,
+        m.last_message_at
-        r.trust_score
       FROM (
         SELECT
           user_id,
@@ -170,7 +168,6 @@ export class DashboardRepository {
         GROUP BY user_id, username, avatar_url
       ) m
       LEFT JOIN ${pgUserProfilesTable} p ON p.user_id = m.user_id
-      LEFT JOIN ${pgUserReputationsTable} r ON r.user_id = m.user_id
       ${whereClause}
       ORDER BY m.last_message_at DESC NULLS LAST
       LIMIT ${limit + 1}
@@ -186,10 +183,6 @@ export class DashboardRepository {
         total_messages: Number(r.total_messages),
         flagged_count: Number(r.flagged_count),
         last_message_at: r.last_message_at ? Number(r.last_message_at) : null,
-        trust_score:
-          r.trust_score !== null && r.trust_score !== undefined
-            ? Number(r.trust_score)
-            : null,
       }));
 
     const lastRow = rows[limit - 1] as Record<string, unknown> | undefined;
@@ -437,10 +430,7 @@ export class DashboardRepository {
         m.flagged_count,
         m.clean_count,
         p.profile_summary,
-        p.last_analyzed_at,
+        p.last_analyzed_at
-        r.trust_score,
-        r.clean_message_streak,
-        r.total_infractions
       FROM (
         SELECT
           user_id,
@@ -454,7 +444,6 @@ export class DashboardRepository {
         GROUP BY user_id, username, avatar_url
       ) m
       LEFT JOIN ${pgUserProfilesTable} p ON p.user_id = m.user_id
-      LEFT JOIN ${pgUserReputationsTable} r ON r.user_id = m.user_id
     `);
 
     const row = userResult.rows[0] as Record<string, unknown> | undefined;
@@ -481,13 +470,6 @@ export class DashboardRepository {
       last_analyzed_at: row.last_analyzed_at
         ? Number(row.last_analyzed_at)
         : null,
-      trust_score: row.trust_score != null ? Number(row.trust_score) : null,
-      clean_message_streak:
-        row.clean_message_streak != null
-          ? Number(row.clean_message_streak)
-          : null,
-      total_infractions:
-        row.total_infractions != null ? Number(row.total_infractions) : null,
       recent_messages: (recent.rows as Record<string, unknown>[]).map((r) => ({
         id: String(r.id),
         content: String(r.content),

services/backend/src/modules/dashboard/dashboard.routes.ts

@@ -1,111 +0,0 @@
-import type { Request, Response, Router } from "express";
-import express from "express";
-import { createChildLogger } from "@/shared/logger/index";
-import { asyncHandler } from "../../shared/middlewares/index.js";
-import { dashboardService } from "./dashboard.service.js";
-
-const logger = createChildLogger("dashboard.routes");
-
-export function createDashboardRouter(): Router {
-  const router = express.Router();
-
-  // GET /api/dashboard/stats — aggregated server statistics
-  router.get(
-    "/dashboard/stats",
-    asyncHandler(async (_req: Request, res: Response) => {
-      logger.debug("Fetching dashboard stats");
-      const stats = await dashboardService.getStats();
-      res.json(stats);
-    }),
-  );
-
-  // GET /api/dashboard/activity?days=14 — message volume over time
-  router.get(
-    "/dashboard/activity",
-    asyncHandler(async (req: Request, res: Response) => {
-      const days = Math.min(Math.max(Number(req.query.days) || 14, 1), 90);
-      const activity = await dashboardService.getActivity(days);
-      res.json(activity);
-    }),
-  );
-
-  // GET /api/dashboard/users — paginated user list with profiles
-  router.get(
-    "/dashboard/users",
-    asyncHandler(async (req: Request, res: Response) => {
-      const limit = Number(req.query.limit) || 20;
-      const cursor =
-        typeof req.query.cursor === "string" ? req.query.cursor : undefined;
-      const search =
-        typeof req.query.search === "string" ? req.query.search : undefined;
-
-      const result = await dashboardService.listUsers({
-        limit,
-        cursor,
-        search,
-      });
-      res.json(result);
-    }),
-  );
-
-  // GET /api/dashboard/users/:userId — single user detail
-  router.get(
-    "/dashboard/users/:userId",
-    asyncHandler(async (req: Request, res: Response) => {
-      const userId = String(req.params.userId);
-      const detail = await dashboardService.getUserDetail(userId);
-      res.json(detail);
-    }),
-  );
-
-  // GET /api/dashboard/channels — paginated channel list with culture summaries
-  router.get(
-    "/dashboard/channels",
-    asyncHandler(async (req: Request, res: Response) => {
-      const limit = Number(req.query.limit) || 20;
-      const search =
-        typeof req.query.search === "string" ? req.query.search : undefined;
-      const guildId =
-        typeof req.query.guild_id === "string" ? req.query.guild_id : undefined;
-
-      const result = await dashboardService.listChannels({
-        limit,
-        search,
-        guildId,
-      });
-      res.json(result);
-    }),
-  );
-
-  // GET /api/dashboard/channels/:channelId — single channel detail
-  router.get(
-    "/dashboard/channels/:channelId",
-    asyncHandler(async (req: Request, res: Response) => {
-      const channelId = String(req.params.channelId);
-      const detail = await dashboardService.getChannelDetail(channelId);
-      res.json(detail);
-    }),
-  );
-
-  // GET /api/dashboard/reactions — top reacted messages
-  router.get(
-    "/dashboard/reactions",
-    asyncHandler(async (req: Request, res: Response) => {
-      const limit = Number(req.query.limit) || 20;
-      const reactions = await dashboardService.getTopReactions(limit);
-      res.json(reactions);
-    }),
-  );
-
-  // GET /api/dashboard/reactors — top users by reactions given
-  router.get(
-    "/dashboard/reactors",
-    asyncHandler(async (req: Request, res: Response) => {
-      const limit = Number(req.query.limit) || 20;
-      const reactors = await dashboardService.getTopReactors(limit);
-      res.json(reactors);
-    }),
-  );
-
-  return router;
-}

services/backend/src/modules/dashboard/index.ts

@@ -1 +0,0 @@
-export { createDashboardRouter } from "./dashboard.routes.js";

services/backend/src/modules/health/moderationMetrics.ts

@@ -67,9 +67,9 @@ export const moderationErrors = new Counter({
   labelNames: ["type"] as const,
 });
 
-export const searxngCalls = new Counter({
+export const webSearchCalls = new Counter({
-  name: "moderation_searxng_calls_total",
+  name: "moderation_websearch_calls_total",
-  help: "SearXNG search calls",
+  help: "Wikipedia web-search calls",
   labelNames: ["status"] as const,
 });
 

services/backend/src/modules/knowledge/knowledge.repository.ts

@@ -0,0 +1,99 @@
+import { sql } from "drizzle-orm";
+import { getDatabase } from "../../shared/database/index.js";
+
+export interface ChannelCultureRow {
+  channel_id: string;
+  guild_id: string | null;
+  channel_name: string | null;
+  culture_summary: string | null;
+  last_analyzed_at: number | null;
+}
+
+export interface GlossaryRow {
+  term: string;
+  definition: string;
+  source_url: string;
+  resolved_at: number;
+  hit_count: number;
+}
+
+export interface EditHistoryRow {
+  id: string;
+  message_id: string;
+  old_content: string;
+  edited_at: number;
+  channel_id: string | null;
+  channel_name: string | null;
+  username: string | null;
+}
+
+export class KnowledgeRepository {
+  /** Public read-only channel culture glossary (AI-generated norms/slang). */
+  async listChannelCultures(limit = 50, search?: string) {
+    const db = getDatabase();
+    const conditions: string[] = [];
+    if (search) {
+      conditions.push(
+        `(c.channel_id ILIKE '%${search.replace(/'/g, "''")}%' OR c.culture_summary ILIKE '%${search.replace(/'/g, "''")}%')`,
+      );
+    }
+    const where = conditions.length ? `WHERE ${conditions.join(" AND ")}` : "";
+    const result = await db.execute(
+      sql.raw(`
+        SELECT
+          c.channel_id,
+          c.guild_id,
+          COALESCE(NULLIF((
+            SELECT (metadata::jsonb -> 'channel' ->> 'channelName')
+            FROM messages WHERE channel_id = c.channel_id AND metadata IS NOT NULL
+            LIMIT 1
+          ), ''), c.channel_id) AS channel_name,
+          c.culture_summary,
+          c.last_analyzed_at
+        FROM channel_cultures c
+        ${where}
+        ORDER BY c.last_analyzed_at DESC NULLS LAST
+        LIMIT ${limit}
+      `),
+    );
+    const rows = (result.rows as Record<string, unknown>[]) || [];
+    return rows.map((r) => ({
+      channel_id: String(r.channel_id),
+      guild_id: r.guild_id ? String(r.guild_id) : null,
+      channel_name: r.channel_name ? String(r.channel_name) : null,
+      culture_summary: r.culture_summary ? String(r.culture_summary) : null,
+      last_analyzed_at: r.last_analyzed_at ? Number(r.last_analyzed_at) : null,
+    }));
+  }
+
+  /** Public read-only term knowledge base (resolved via Wikipedia/SearXNG). */
+  async listGlossary(limit = 50, search?: string) {
+    const db = getDatabase();
+    const conditions: string[] = [];
+    if (search) {
+      conditions.push(
+        `(term ILIKE '%${search.replace(/'/g, "''")}%' OR definition ILIKE '%${search.replace(/'/g, "''")}%')`,
+      );
+    }
+    const where = conditions.length ? `WHERE ${conditions.join(" AND ")}` : "";
+    const result = await db.execute(
+      sql.raw(`
+        SELECT term, definition, source_url, resolved_at, hit_count
+        FROM term_glossary_cache
+        ${where}
+        ORDER BY hit_count DESC, resolved_at DESC
+        LIMIT ${limit}
+      `),
+    );
+    const rows = (result.rows as Record<string, unknown>[]) || [];
+    return rows.map((r) => ({
+      term: String(r.term),
+      definition: String(r.definition ?? ""),
+      source_url: r.source_url ? String(r.source_url) : "",
+      resolved_at: r.resolved_at ? Number(r.resolved_at) : 0,
+      hit_count: Number(r.hit_count ?? 0),
+    }));
+  }
+}
+
+export const knowledgeRepository = new KnowledgeRepository();

services/backend/src/modules/knowledge/knowledge.service.ts

@@ -0,0 +1,18 @@
+import { createChildLogger } from "../../shared/logger/index.js";
+import { knowledgeRepository } from "./knowledge.repository.js";
+
+const logger = createChildLogger("knowledge.service");
+
+export class KnowledgeService {
+  async listChannelCultures(limit = 50, search?: string) {
+    logger.debug({ limit, search }, "Listing channel cultures");
+    return knowledgeRepository.listChannelCultures(limit, search);
+  }
+
+  async listGlossary(limit = 50, search?: string) {
+    logger.debug({ limit, search }, "Listing glossary terms");
+    return knowledgeRepository.listGlossary(limit, search);
+  }
+}
+
+export const knowledgeService = new KnowledgeService();

services/backend/src/modules/media/index.ts

@@ -1 +0,0 @@
-export { createMediaRouter } from "./media.routes.js";

services/backend/src/modules/media/media.routes.ts

@@ -1,71 +0,0 @@
-import type { Request, Response, Router } from "express";
-import express from "express";
-import { createChildLogger } from "@/shared/logger/index";
-import { asyncHandler, validateBody } from "../../shared/middlewares/index.js";
-import { mediaLoopSchema, mediaQueueSchema } from "./media.schema.js";
-import { getStatus, queue, setLoop, skip, stop } from "./media.service.js";
-
-const logger = createChildLogger("media.routes");
-
-export function createMediaRouter(): Router {
-  const router = express.Router();
-
-  // GET /api/media/status
-  router.get(
-    "/media/status",
-    asyncHandler(async (_req: Request, res: Response) => {
-      logger.debug("Media status requested");
-      const status = await getStatus();
-      res.json(status);
-    }),
-  );
-
-  // POST /api/media/queue
-  router.post(
-    "/media/queue",
-    validateBody(mediaQueueSchema),
-    asyncHandler(async (req: Request, res: Response) => {
-      const { source, mode } = req.body as {
-        source: string;
-        mode: "music" | "screen";
-      };
-      logger.debug({ source, mode }, "Media queue requested");
-      const state = await queue(source, mode);
-      res.json(state);
-    }),
-  );
-
-  // POST /api/media/skip
-  router.post(
-    "/media/skip",
-    asyncHandler(async (_req: Request, res: Response) => {
-      logger.debug("Media skip requested");
-      const state = await skip();
-      res.json(state);
-    }),
-  );
-
-  // POST /api/media/stop
-  router.post(
-    "/media/stop",
-    asyncHandler(async (_req: Request, res: Response) => {
-      logger.debug("Media stop requested");
-      const state = await stop();
-      res.json(state);
-    }),
-  );
-
-  // POST /api/media/loop
-  router.post(
-    "/media/loop",
-    validateBody(mediaLoopSchema),
-    asyncHandler(async (req: Request, res: Response) => {
-      const { loop } = req.body as { loop: boolean };
-      logger.debug({ loop }, "Media loop requested");
-      const state = await setLoop(loop);
-      res.json(state);
-    }),
-  );
-
-  return router;
-}

services/backend/src/modules/messages/embed.ts

@@ -0,0 +1,43 @@
+import { config } from "@/shared/config/index";
+import { createChildLogger } from "@/shared/logger/index";
+
+const logger = createChildLogger("messages-embed");
+
+/**
+ * Embed a search query with the configured OpenAI-compatible embedding model.
+ * Uses raw fetch (the backend has no openai SDK dependency) and returns null
+ * when embeddings are not configured (search unavailable).
+ *
+ * encoding_format: "float" is REQUIRED — Nvidia-backed models reject base64.
+ */
+export async function embedQuery(text: string): Promise<number[] | null> {
+  if (!config.AI_LLM_API_KEY || !config.AI_LLM_EMBEDDING_MODEL) return null;
+  try {
+    const res = await fetch(`${config.AI_LLM_BASE_URL}/embeddings`, {
+      method: "POST",
+      headers: {
+        "Content-Type": "application/json",
+        Authorization: `Bearer ${config.AI_LLM_API_KEY}`,
+      },
+      body: JSON.stringify({
+        model: config.AI_LLM_EMBEDDING_MODEL,
+        input: text,
+        encoding_format: "float",
+      }),
+    });
+    if (!res.ok) {
+      logger.warn({ status: res.status }, "query embed HTTP error");
+      return null;
+    }
+    const json = (await res.json()) as {
+      data?: Array<{ embedding?: number[] }>;
+    };
+    return json.data?.[0]?.embedding ?? null;
+  } catch (error) {
+    logger.warn(
+      { error: error instanceof Error ? error.message : String(error) },
+      "query embed failed",
+    );
+    return null;
+  }
+}

services/backend/src/modules/messages/index.ts

@@ -1 +0,0 @@
-export { createMessagesRouter } from "./messages.routes.js";

services/backend/src/modules/messages/messages.controller.ts

@@ -1,74 +0,0 @@
-import type { Request, Response } from "express";
-import { createChildLogger } from "@/shared/logger/index";
-import { asyncHandler } from "../../shared/middlewares/index.js";
-import { messageQuerySchema } from "./messages.schema.js";
-import { messagesService } from "./messages.service.js";
-
-const logger = createChildLogger("messages.controller");
-
-export const handleListMessages = asyncHandler(
-  async (req: Request, res: Response) => {
-    const query = messageQuerySchema.parse(req.query);
-    logger.debug({ query }, "Handling list messages request");
-    const result = await messagesService.listMessages(query);
-    res.json(result);
-  },
-);
-
-export const handleGetMessagesByChannel = asyncHandler(
-  async (req: Request, res: Response) => {
-    if (!req.params.channelId) {
-      res.status(400).json({ error: "Missing route parameter: channelId" });
-      return;
-    }
-    const channelId = req.params.channelId as string;
-    const query = messageQuerySchema.parse(req.query);
-    logger.debug({ channelId, query }, "Handling get messages by channel");
-    const result = await messagesService.getMessagesByChannel(channelId, query);
-    res.json(result);
-  },
-);
-
-export const handleGetMessageById = asyncHandler(
-  async (req: Request, res: Response) => {
-    if (!req.params.id) {
-      res.status(400).json({ error: "Missing route parameter: id" });
-      return;
-    }
-    const id = req.params.id as string;
-    logger.debug({ id }, "Handling get message by ID");
-    const result = await messagesService.getMessageById(id);
-    res.json(result);
-  },
-);
-
-export const handleGetImageMessages = asyncHandler(
-  async (req: Request, res: Response) => {
-    const guildId = req.query.guildId as string | undefined;
-    if (!guildId) {
-      res.status(400).json({ error: "Missing query parameter: guildId" });
-      return;
-    }
-    const limit = Number(req.query.limit) || 50;
-    logger.debug({ guildId, limit }, "Handling get image messages");
-    const result = await messagesService.getImageMessages(guildId, limit);
-    res.json(result);
-  },
-);
-
-export const handleGetAttachmentsByChannel = asyncHandler(
-  async (req: Request, res: Response) => {
-    if (!req.params.channelId) {
-      res.status(400).json({ error: "Missing route parameter: channelId" });
-      return;
-    }
-    const channelId = req.params.channelId as string;
-    const query = messageQuerySchema.parse(req.query);
-    logger.debug({ channelId, query }, "Handling get attachments by channel");
-    const result = await messagesService.getAttachmentsByChannel(
-      channelId,
-      query,
-    );
-    res.json(result);
-  },
-);

services/backend/src/modules/messages/messages.repository.ts

@@ -172,6 +172,71 @@ export class MessagesRepository {
     return { data, nextCursor };
   }
 
+  /**
+   * Async generator that yields messages ONE AT A TIME for WS streaming.
+   * Each `.next()` runs its own bounded DB query (limit+1) advancing on the
+   * `created_at` cursor, so memory stays flat and the caller can emit one WS
+   * frame per message (no 50-row batch). Stops when a page returns < limit.
+   */
+  async *streamMany(
+    query: MessageQuery,
+    pageSize = 50,
+  ): AsyncGenerator<ReturnType<typeof mapMessageRow>, void, unknown> {
+    const conditions: SQL[] = [];
+
+    if (query.guildId) {
+      conditions.push(eq(pgMessagesTable.guild_id, query.guildId));
+    }
+    if (query.channelId) {
+      conditions.push(eq(pgMessagesTable.channel_id, query.channelId));
+    }
+    if (query.userId) {
+      conditions.push(eq(pgMessagesTable.user_id, query.userId));
+    }
+    if (query.status) {
+      conditions.push(eq(pgMessagesTable.ai_status, query.status));
+    }
+    if (EXCLUDED_THREAD_IDS.length > 0) {
+      const excludeThreads = or(
+        isNull(pgMessagesTable.thread_id),
+        notInArray(pgMessagesTable.thread_id, EXCLUDED_THREAD_IDS),
+      );
+      if (excludeThreads) conditions.push(excludeThreads);
+    }
+
+    const where = conditions.length > 0 ? and(...conditions) : undefined;
+    let cursor: string | undefined = query.cursor;
+
+    while (true) {
+      const pageConditions = where ? [where] : [];
+      if (cursor) {
+        pageConditions.push(lt(pgMessagesTable.created_at, Number(cursor)));
+      }
+      const pageWhere =
+        pageConditions.length > 0 ? and(...pageConditions) : undefined;
+
+      const db = getDatabase();
+      const rows = await db
+        .select()
+        .from(pgMessagesTable)
+        .where(pageWhere)
+        .orderBy(desc(pgMessagesTable.created_at))
+        .limit(pageSize + 1);
+
+      if (rows.length === 0) return;
+
+      const hasMore = rows.length > pageSize;
+      const pageRows = hasMore ? rows.slice(0, pageSize) : rows;
+
+      for (const r of pageRows) {
+        yield mapMessageRow(r as Record<string, unknown>);
+      }
+
+      if (!hasMore) return;
+      cursor = String(rows[pageSize - 1].created_at);
+    }
+  }
+
   async create(data: MessageCreate) {
     const db = getDatabase();
     const id = crypto.randomUUID();
@@ -394,6 +459,69 @@ export class MessagesRepository {
 
     return { data: trimmed, nextCursor };
   }
+
+  /**
+   * Per-hour message volume for the last `days` days, grouped by channel.
+   * Powers the public Activity Heatmap (read-only, no write scope).
+   * Returns a flat list of { channel_id, hour (0-23), count } buckets.
+   */
+  async getActivity(days = 30) {
+    const db = getDatabase();
+    const since = Date.now() - days * 24 * 60 * 60 * 1000;
+    const result = await db.execute(sql`
+      SELECT channel_id,
+             EXTRACT(HOUR FROM to_timestamp(created_at / 1000))::int AS hour,
+             COUNT(*)::int AS c
+      FROM messages
+      WHERE created_at >= ${since}
+      GROUP BY channel_id, hour
+      ORDER BY channel_id, hour
+    `);
+    const rows = (result.rows as Record<string, unknown>[]) || [];
+    return rows.map((r) => ({
+      channelId: String(r.channel_id ?? "unknown"),
+      hour: Number(r.hour ?? 0),
+      count: Number(r.c ?? 0),
+    }));
+  }
+
+  /**
+   * Recent message edits across the server (evasion-signal tracker).
+   * Public, read-only. Joins message_edits → messages for context.
+   */
+  async getRecentEdits(limit = 50, channelId?: string) {
+    const db = getDatabase();
+    const where = channelId
+      ? `WHERE m.channel_id = '${channelId.replace(/'/g, "''")}'`
+      : "";
+    const result = await db.execute(
+      sql.raw(`
+        SELECT
+          e.id,
+          e.message_id,
+          e.old_content,
+          e.edited_at,
+          m.channel_id,
+          COALESCE(NULLIF((m.metadata::jsonb -> 'channel' ->> 'channelName'), ''), m.channel_id) AS channel_name,
+          m.username
+        FROM message_edits e
+        JOIN messages m ON m.id = e.message_id
+        ${where}
+        ORDER BY e.edited_at DESC
+        LIMIT ${limit}
+      `),
+    );
+    const rows = (result.rows as Record<string, unknown>[]) || [];
+    return rows.map((r) => ({
+      id: String(r.id),
+      message_id: String(r.message_id),
+      old_content: r.old_content ? String(r.old_content) : "",
+      edited_at: r.edited_at ? Number(r.edited_at) : 0,
+      channel_id: r.channel_id ? String(r.channel_id) : null,
+      channel_name: r.channel_name ? String(r.channel_name) : null,
+      username: r.username ? String(r.username) : null,
+    }));
+  }
 }
 
 export const messagesRepository = new MessagesRepository();

services/backend/src/modules/messages/messages.routes.ts

@@ -1,51 +0,0 @@
-import type { Request, Response, Router } from "express";
-import express from "express";
-import { createChildLogger } from "@/shared/logger/index";
-import { asyncHandler } from "../../shared/middlewares/index.js";
-import {
-  handleGetAttachmentsByChannel,
-  handleGetImageMessages,
-  handleGetMessageById,
-  handleGetMessagesByChannel,
-  handleListMessages,
-} from "./messages.controller.js";
-import { messagesService } from "./messages.service.js";
-
-const logger = createChildLogger("messages.routes");
-
-export function createMessagesRouter(): Router {
-  const router = express.Router();
-
-  // GET /api/messages/images - Get messages with image attachments
-  // MUST be registered BEFORE /messages/:channelId so "images" is not
-  // captured as a channelId param.
-  router.get("/messages/images", handleGetImageMessages);
-
-  // GET /api/messages - List messages
-  router.get("/messages", handleListMessages);
-
-  // GET /api/messages/:channelId - Get messages by channel
-  router.get("/messages/:channelId", handleGetMessagesByChannel);
-
-  // GET /api/messages/:channelId/attachments - Get attachments by channel
-  router.get("/messages/:channelId/attachments", handleGetAttachmentsByChannel);
-
-  // GET /api/messages/detail/:id - Get single message by ID
-  // (uses /detail/ prefix to avoid collision with :channelId route above)
-  router.get("/messages/detail/:id", handleGetMessageById);
-
-  // GET /api/review - Get flagged/warned messages for review
-  router.get(
-    "/review",
-    asyncHandler(async (req: Request, res: Response) => {
-      const limit = Number(req.query.limit) || 20;
-      const channelId = (req.query.channelId as string) || undefined;
-
-      const rows = await messagesService.getReviewMessages(channelId, limit);
-      logger.debug({ limit, channelId }, "Review query executed");
-      res.json({ results: rows, limit, cursor: null });
-    }),
-  );
-
-  return router;
-}

services/backend/src/modules/messages/messages.schema.ts

@@ -41,3 +41,11 @@ export const messageUpdateSchema = z.object({
 export type MessageQuery = z.infer<typeof messageQuerySchema>;
 export type MessageCreate = z.infer<typeof messageCreateSchema>;
 export type MessageUpdate = z.infer<typeof messageUpdateSchema>;
+
+export const semanticSearchSchema = z.object({
+  query: z.string().min(1).max(500),
+  limit: z.coerce.number().int().positive().max(50).default(10),
+  guildId: z.string().optional(),
+});
+
+export type SemanticSearchQuery = z.infer<typeof semanticSearchSchema>;

services/backend/src/modules/messages/messages.service.ts

@@ -1,7 +1,9 @@
 import { NotFoundError, ValidationError } from "@/shared/errors/index";
 import { createChildLogger } from "@/shared/logger/index";
+import { embedQuery } from "./embed.js";
 import { messagesRepository } from "./messages.repository.js";
-import type { MessageQuery } from "./messages.schema.js";
+import type { MessageQuery, SemanticSearchQuery } from "./messages.schema.js";
+import { searchArchive } from "./qdrant.js";
 
 const logger = createChildLogger("messages.service");
 
@@ -15,6 +17,14 @@ export class MessagesService {
     return messagesRepository.findMany(query);
   }
 
+  /**
+   * Stream messages one at a time (no 50-row batch). The WS handler iterates
+   * this generator and emits one `message_snapshot` frame per message.
+   */
+  streamMessages(query: MessageQuery, pageSize = 50) {
+    return messagesRepository.streamMany(query, pageSize);
+  }
+
   async getMessagesByChannel(channelId: string, query: MessageQuery) {
     if (!channelId) {
       throw new ValidationError("channelId is required");
@@ -70,6 +80,49 @@ export class MessagesService {
     logger.debug({ channelId, limit }, "Getting review messages");
     return messagesRepository.getReviewMessages(channelId, limit);
   }
+
+  /**
+   * Public, read-only semantic search over the persistent message archive.
+   * Embeds the query, searches Qdrant, returns text + metadata. Best-effort:
+   * if embeddings/Qdrant are unavailable, returns an empty result set.
+   */
+  async semanticSearch(
+    input: SemanticSearchQuery,
+  ): Promise<{ results: ReturnType<typeof mapSearchHit>[]; nextCursor: null }> {
+    const vector = await embedQuery(input.query);
+    if (!vector) {
+      logger.debug(
+        { query: input.query },
+        "semantic search skipped: no embedder",
+      );
+      return { results: [], nextCursor: null };
+    }
+    const hits = await searchArchive(vector, input.limit, 0.6);
+    const results = hits.map((h) => mapSearchHit(h));
+    return { results, nextCursor: null };
+  }
+
+  async getActivity(days = 30) {
+    return messagesRepository.getActivity(days);
+  }
+
+  async getRecentEdits(limit = 50, channelId?: string) {
+    logger.debug({ limit, channelId }, "Getting recent message edits");
+    return messagesRepository.getRecentEdits(limit, channelId);
+  }
+}
+
+/** Shape returned to the frontend (text + metadata from the archive payload). */
+function mapSearchHit(hit: {
+  score: number;
+  payload: { text: string; content_hash?: string; analyzed_at: number };
+}) {
+  return {
+    message_id: hit.payload.content_hash ?? null,
+    content: hit.payload.text,
+    score: hit.score,
+    created_at: hit.payload.analyzed_at,
+  };
 }
 
 export const messagesService = new MessagesService();

services/backend/src/modules/messages/qdrant.ts

@@ -0,0 +1,95 @@
+import { config } from "@/shared/config/index";
+import { createChildLogger } from "@/shared/logger/index";
+
+const logger = createChildLogger("messages-qdrant");
+
+export interface ArchiveHit {
+  score: number;
+  payload: {
+    text: string;
+    content_hash?: string;
+    analyzed_at: number;
+    expires_at: number;
+  };
+}
+
+function baseUrl(): string {
+  return (config.QDRANT_URL ?? "http://100.121.180.82:6333").replace(
+    /\/+$/,
+    "",
+  );
+}
+
+function headers(): Record<string, string> {
+  const h: Record<string, string> = { "Content-Type": "application/json" };
+  if (config.QDRANT_API_KEY) h["api-key"] = config.QDRANT_API_KEY;
+  return h;
+}
+
+export const ARCHIVE_COLLECTION =
+  config.QDRANT_ARCHIVE_COLLECTION ?? "gmw_message_archive";
+
+async function request(
+  method: string,
+  path: string,
+  body?: unknown,
+  timeoutMs = 10_000,
+): Promise<unknown> {
+  const controller = new AbortController();
+  const timer = setTimeout(() => controller.abort(), timeoutMs);
+  try {
+    const res = await fetch(`${baseUrl()}${path}`, {
+      method,
+      headers: headers(),
+      body: body === undefined ? undefined : JSON.stringify(body),
+      signal: controller.signal,
+    });
+    const text = await res.text();
+    if (!res.ok) {
+      throw new Error(
+        `Qdrant ${method} ${path} -> ${res.status}: ${text.slice(0, 200)}`,
+      );
+    }
+    return text ? JSON.parse(text) : null;
+  } finally {
+    clearTimeout(timer);
+  }
+}
+
+/** Search the archive collection for the nearest vectors to `vector`. */
+export async function searchArchive(
+  vector: number[],
+  limit: number,
+  scoreThreshold: number,
+): Promise<ArchiveHit[]> {
+  if (!config.QDRANT_URL) return [];
+  try {
+    const json = (await request(
+      "POST",
+      `/collections/${ARCHIVE_COLLECTION}/points/search`,
+      {
+        vector,
+        limit,
+        score_threshold: scoreThreshold,
+        with_payload: true,
+      },
+    )) as {
+      result?: Array<{
+        score?: number;
+        payload?: ArchiveHit["payload"];
+      }>;
+    };
+    return (json.result ?? [])
+      .filter((h) => h.payload?.text)
+      .map((h) => ({
+        score: h.score ?? 0,
+        payload: h.payload as ArchiveHit["payload"],
+      }));
+  } catch (error) {
+    logger.warn(
+      { error: error instanceof Error ? error.message : String(error) },
+      "archive search failed",
+    );
+    return [];
+  }
+}

services/backend/src/modules/moderation/index.ts

@@ -1 +0,0 @@
-export { createModerationRouter } from "./moderation.routes.js";

services/backend/src/modules/moderation/moderation.repository.ts

@@ -17,6 +17,20 @@ const ACTION_TYPES = [
 ] as const;
 const STATUSES = ["pending", "executed", "failed"] as const;
 
+/** Parse a JSON-stringified array column (e.g. flags/categories/evidence).
+ *  Returns null on empty/malformed input so the FE can treat it as "no data". */
+function parseJsonArray(value: unknown): string[] | null {
+  if (value == null) return null;
+  const str = typeof value === "string" ? value : String(value);
+  if (str.length === 0) return null;
+  try {
+    const parsed = JSON.parse(str);
+    return Array.isArray(parsed) ? (parsed as string[]) : null;
+  } catch {
+    return null;
+  }
+}
+
 export class ModerationRepository {
   async getStats() {
     const db = getDatabase();
@@ -103,6 +117,13 @@ export class ModerationRepository {
         a.error,
         a.created_at,
         a.executed_at,
+        a.flags,
+        a.categories,
+        a.severity,
+        a.confidence,
+        a.score,
+        a.evidence,
+        a.policy_version,
         m.username,
         LEFT(m.content, 300) AS content
       FROM moderation_actions a
@@ -126,6 +147,13 @@ export class ModerationRepository {
       error: r.error ? String(r.error) : null,
       created_at: r.created_at ? Number(r.created_at) : null,
       executed_at: r.executed_at ? Number(r.executed_at) : null,
+      flags: parseJsonArray(r.flags),
+      categories: parseJsonArray(r.categories),
+      severity: r.severity ? String(r.severity) : null,
+      confidence: r.confidence != null ? Number(r.confidence) : null,
+      score: r.score != null ? Number(r.score) : null,
+      evidence: parseJsonArray(r.evidence),
+      policy_version: r.policy_version ? String(r.policy_version) : null,
       username: r.username ? String(r.username) : null,
       content: r.content ? String(r.content) : null,
     }));
@@ -136,6 +164,220 @@ export class ModerationRepository {
 
     return { data, nextCursor };
   }
+
+  /**
+   * Aggregate moderation trends over the last `days` days.
+   * - category counts (from the jsonb/text[] `categories` column, unnested)
+   * - severity distribution
+   * - action_type distribution
+   * Read-only; powers the public Toxic Topic Trends panel.
+   */
+  async getTrends(days: number) {
+    const db = getDatabase();
+    const since = Date.now() - days * 24 * 60 * 60 * 1000;
+
+    const cats = await db.execute(sql`
+      SELECT jsonb_array_elements_text(a.categories::jsonb) AS cat, COUNT(*)::int AS c
+      FROM moderation_actions a
+      WHERE a.created_at >= ${since} AND a.categories IS NOT NULL AND a.categories != '[]' AND a.categories != ''
+      GROUP BY cat
+      ORDER BY c DESC
+      LIMIT 15
+    `);
+    const catRows = (cats.rows as Record<string, unknown>[]) || [];
+
+    const sev = await db.execute(sql`
+      SELECT severity, COUNT(*)::int AS c
+      FROM moderation_actions
+      WHERE created_at >= ${since} AND severity IS NOT NULL
+      GROUP BY severity
+    `);
+    const sevRows = (sev.rows as Record<string, unknown>[]) || [];
+
+    const act = await db.execute(sql`
+      SELECT action_type, COUNT(*)::int AS c
+      FROM moderation_actions
+      WHERE created_at >= ${since}
+      GROUP BY action_type
+      ORDER BY c DESC
+    `);
+    const actRows = (act.rows as Record<string, unknown>[]) || [];
+
+    return {
+      categories: catRows.map((r) => ({
+        name: String(r.cat),
+        count: Number(r.c ?? 0),
+      })),
+      severities: sevRows.map((r) => ({
+        level: String(r.severity),
+        count: Number(r.c ?? 0),
+      })),
+      actions: actRows.map((r) => ({
+        type: String(r.action_type),
+        count: Number(r.c ?? 0),
+      })),
+    };
+  }
+
+  /**
+   * Top flagged domains over the last `days` days.
+   * Extracts the host from any URL in `content`/`reason`/`evidence` and ranks
+   * by how often it appears in moderation actions. Powers the Scam Domain panel.
+   */
+  async getTopFlaggedDomains(days: number) {
+    const db = getDatabase();
+    const since = Date.now() - days * 24 * 60 * 60 * 1000;
+    const result = await db.execute(sql`
+      SELECT host, COUNT(*)::int AS c
+      FROM (
+        SELECT DISTINCT a.id,
+          (regexp_matches(COALESCE(a.content,'') || ' ' || COALESCE(a.reason,'') || ' ' || COALESCE(a.evidence,''), 'https?://([^/\s?#]+)', 'g'))[1] AS host
+        FROM moderation_actions a
+        WHERE a.created_at >= ${since}
+          AND (a.content IS NOT NULL OR a.reason IS NOT NULL OR a.evidence IS NOT NULL)
+      ) sub
+      WHERE host IS NOT NULL
+      GROUP BY host
+      ORDER BY c DESC
+      LIMIT 20
+    `);
+    const rows = (result.rows as Record<string, unknown>[]) || [];
+    return rows.map((r) => ({
+      domain: String(r.host).toLowerCase(),
+      count: Number(r.c ?? 0),
+    }));
+  }
+
+  /**
+   * Top flagged channels over the last `days` days.
+   * Joins moderation_actions → messages to attribute each action to a channel.
+   * Powers the Top Flagged Channels panel.
+   */
+  async getTopFlaggedChannels(days: number) {
+    const db = getDatabase();
+    const since = Date.now() - days * 24 * 60 * 60 * 1000;
+    const result = await db.execute(sql`
+      SELECT
+        m.channel_id,
+        COALESCE(NULLIF((m.metadata::jsonb -> 'channel' ->> 'channelName'), ''), m.channel_id) AS channel_name,
+        COUNT(*)::int AS flagged_count
+      FROM moderation_actions a
+      LEFT JOIN messages m ON m.id = a.message_id
+      WHERE a.created_at >= ${since} AND m.channel_id IS NOT NULL
+      GROUP BY m.channel_id, (m.metadata::jsonb -> 'channel' ->> 'channelName')
+      ORDER BY flagged_count DESC
+      LIMIT 15
+    `);
+    const rows = (result.rows as Record<string, unknown>[]) || [];
+    return rows.map((r) => ({
+      channel_id: String(r.channel_id),
+      channel_name: r.channel_name ? String(r.channel_name) : null,
+      flagged_count: Number(r.flagged_count),
+    }));
+  }
+
+  /**
+   * Hour-of-day distribution of moderation actions over the last `days` days.
+   * 24 rows (hour 0..23), with total + flagged-by-severity counts.
+   * Powers the Moderation Heatmap by Hour panel.
+   */
+  async getHourlyModeration(days: number) {
+    const db = getDatabase();
+    const since = Date.now() - days * 24 * 60 * 60 * 1000;
+    const result = await db.execute(sql`
+      SELECT
+        EXTRACT(HOUR FROM to_timestamp(created_at / 1000))::int AS hour,
+        COUNT(*)::int AS total
+      FROM moderation_actions
+      WHERE created_at >= ${since}
+      GROUP BY hour
+      ORDER BY hour
+    `);
+    const rows = (result.rows as Record<string, unknown>[]) || [];
+    const byHour = new Map<number, number>();
+    for (const r of rows) byHour.set(Number(r.hour), Number(r.total));
+    return Array.from({ length: 24 }, (_, h) => ({
+      hour: h,
+      total: byHour.get(h) ?? 0,
+    }));
+  }
+
+  /**
+   * Moderation actions filtered to a single category (drill-down).
+   * Powers the Flag Category Drill-down panel.
+   */
+  async getByCategory(days: number, category: string, limit = 50) {
+    const db = getDatabase();
+    const since = Date.now() - days * 24 * 60 * 60 * 1000;
+    const result = await db.execute(
+      sql.raw(`
+      SELECT
+        a.id, a.message_id, a.user_id, a.guild_id, a.action_type,
+        a.reason, a.status, a.created_at, a.severity, a.confidence, a.score,
+        m.username, LEFT(m.content, 300) AS content
+      FROM moderation_actions a
+      LEFT JOIN messages m ON m.id = a.message_id
+      WHERE a.created_at >= ${since}
+        AND a.categories IS NOT NULL
+        AND a.categories::jsonb @> ${JSON.stringify([category])}::jsonb
+      ORDER BY a.created_at DESC
+      LIMIT ${limit}
+    `),
+    );
+    const rows = (result.rows as Record<string, unknown>[]) || [];
+    return rows.map((r) => ({
+      id: String(r.id ?? ""),
+      message_id: r.message_id ? String(r.message_id) : null,
+      user_id: r.user_id ? String(r.user_id) : null,
+      guild_id: String(r.guild_id ?? ""),
+      action_type: String(r.action_type ?? "unknown"),
+      reason: r.reason ? String(r.reason) : null,
+      status: String(r.status ?? "unknown"),
+      created_at: r.created_at ? Number(r.created_at) : null,
+      severity: r.severity ? String(r.severity) : null,
+      confidence: r.confidence != null ? Number(r.confidence) : null,
+      score: r.score != null ? Number(r.score) : null,
+      username: r.username ? String(r.username) : null,
+      content: r.content ? String(r.content) : null,
+    }));
+  }
+
+  /**
+   * Auto-moderation coverage over the last `days` days.
+   * Run completion rate from ai_analysis_runs — what fraction of analysis runs
+   * completed (vs failed/pending). Public "how much is automated" trust metric.
+   */
+  async getCoverage(days: number) {
+    const db = getDatabase();
+    const since = Date.now() - days * 24 * 60 * 60 * 1000;
+    const result = await db.execute(sql`
+      SELECT status, COUNT(*)::int AS c
+      FROM ai_analysis_runs
+      WHERE created_at >= ${since}
+      GROUP BY status
+    `);
+    const rows = (result.rows as Record<string, unknown>[]) || [];
+    const counts: Record<string, number> = {};
+    let total = 0;
+    for (const r of rows) {
+      const s = String(r.status);
+      const c = Number(r.c ?? 0);
+      counts[s] = c;
+      total += c;
+    }
+    const completed = counts.completed ?? 0;
+    const failed = counts.failed ?? 0;
+    const pending = (counts.pending ?? 0) + (counts.processing ?? 0);
+    return {
+      total,
+      completed,
+      failed,
+      pending,
+      coverage_rate:
+        total > 0 ? Number(((completed / total) * 100).toFixed(1)) : 0,
+      failed_rate: total > 0 ? Number(((failed / total) * 100).toFixed(1)) : 0,
+    };
+  }
 }
 
 export const moderationRepository = new ModerationRepository();

services/backend/src/modules/moderation/moderation.routes.ts

@@ -1,43 +0,0 @@
-import type { Request, Response, Router } from "express";
-import express from "express";
-import { createChildLogger } from "../../shared/logger/index.js";
-import { asyncHandler } from "../../shared/middlewares/index.js";
-import { moderationService } from "./moderation.service.js";
-
-const logger = createChildLogger("moderation.routes");
-
-export function createModerationRouter(): Router {
-  const router = express.Router();
-
-  // GET /api/moderation/stats — moderation action summary
-  router.get(
-    "/moderation/stats",
-    asyncHandler(async (_req: Request, res: Response) => {
-      const stats = await moderationService.getStats();
-      res.json(stats);
-    }),
-  );
-
-  // GET /api/moderation/actions — paginated moderation action log
-  router.get(
-    "/moderation/actions",
-    asyncHandler(async (req: Request, res: Response) => {
-      const limit = Number(req.query.limit) || 50;
-      const status = req.query.status as string | undefined;
-      const actionType = req.query.actionType as string | undefined;
-      const cursor = req.query.cursor as string | undefined;
-
-      const result = await moderationService.listActions({
-        limit,
-        status,
-        actionType,
-        cursor: cursor ? Number(cursor) : undefined,
-      });
-
-      logger.debug({ count: result.data.length }, "Moderation actions listed");
-      res.json(result);
-    }),
-  );
-
-  return router;
-}