feat(ai-moderation): rich context + link media vision analysis

- Conversation context recency gates (GAP_MS/MAX_AGE_MS): drop stale
  messages before silence gaps; cold_start anchor + flow descriptor
  tells LLM whether conversation is ongoing or restarted
- [location] block: channel name, thread name, nsfw/age flags from
  captured metadata (thread names instead of bare IDs)
- Link media -> multimodal: text-batch URL fetches that resolve to
  images now run vision analysis (bounded 15s) and switch prompt to
  mixed mode; <web_content> gains og:title for page context
- pnpm-workspace.yaml: approve sharp build script (unblocks install)
This commit is contained in:
asepharyana
2026-08-10 11:26:26 +07:00
parent 5d094829c4
commit 4049ab4201
8 changed files with 497 additions and 24 deletions
@@ -189,6 +189,17 @@ export const configSchema = z
.int()
.positive()
.default(20),
// Recency gates for conversation context. A silence longer than GAP_MS
// between context messages = the conversation restarted (older messages
// dropped); MAX_AGE_MS caps how far back context is considered relevant.
AI_ANALYSIS_CONTEXT_GAP_MS: z.coerce
.number()
.positive()
.default(12 * 60 * 1000),
AI_ANALYSIS_CONTEXT_MAX_AGE_MS: z.coerce
.number()
.positive()
.default(45 * 60 * 1000),
AI_ANALYSIS_PROCESSING_TIMEOUT_MS: z.coerce
.number()
.positive()