The non-streaming Anthropic path leaked OpenAI chat.completion bodies to
Claude clients when the upstream provider forces streaming (forced
SSE->JSON path): parseSSEToOpenAIResponse yields choices[] which Claude
Code cannot parse. Convert to Claude Messages shape for sourceFormat
CLAUDE in sseToJsonHandler.
Shared toClaudeMessageShape/openAICompletionToClaudeMessage moved to
translator/concerns/claudeShape.js to avoid circular import between
nonStreamingHandler and sseToJsonHandler. 5 new tests: sse-to-json
reassembly + Claude guard, plus existing anthropic-nonstream-shape
updated to import from the concern.
BUG: Anthropic client (sourceFormat=CLAUDE) hitting a claude-target provider
got a raw OpenAI chat.completion body on non-streaming requests. The
needsTranslation(CLAUDE,CLAUDE) gate is false when target===source, so the
translator never ran; a claude-transport executor replying OpenAI JSON
(opencode/big-pickle) leaked choices[]/prompt_tokens to the client, which
Anthropic SDKs cannot parse (no content[] blocks, no type:"message").
FIX: shape-aware guard toClaudeMessageShape() in nonStreamingHandler — when
sourceFormat is CLAUDE, convert any OpenAI-shape body to a proper Claude
message (type, content blocks with thinking/text/tool_use, stop_reason via
finish mapping, usage input/output tokens). Claude-shaped bodies pass through.
Also: strip <|im_end|>/<|endoftext|>/<|eot_id|> EOS sentinels from Claude
text deltas in openai-to-claude and kiro-to-claude translators — the upstream
EOS token leaks into the final text_delta (observed 'OK<|im_end|>').
Tests: anthropic-nonstream-shape.test.js (6: eos strip + shape guard incl
tool_calls→tool_use, pass-through, finish mapping). 28/28 related tests green.
New /v1/systemone pass-through route for Jev decision models (jev-1.13,
jev-1.13-free) on OpenCode Zen and the free lane. Follows the media-route
pattern: systemoneConfig in the registry drives URL/headers, the handler
mirrors the embeddings account-fallback + usage flow, and the dashboard
gains a System One media-provider kind. No chat-pipeline changes.
Co-Authored-By: Claude Code <noreply@anthropic.com>
- Match code 110 (billing daily count exceeded) alongside 112/10605/pricingUrl
in isBillingBlock, parsing JSON safely and accepting numeric/string codes
- Accept numeric strings for statusCodeValue and object bodies in envelope peek
- Emit structured 403 quota error chunk instead of synthetic assistant text
when a billing envelope appears mid-stream
- Preserve upstream HTTP status in handleForcedSSEToJson when error chunk carries
a valid 400-599 status
- Add unit tests for code-110 detection, mid-stream billing envelopes, and false-positive guard
Cursor-hosted models (cu/composer-2.5, cu/cursor-grok-*, cu/default) returned
HTTP 200 with an empty turn, or hung, whenever a client sent tools.
- Fold system prompts into the current user message. custom_system_prompt
(RunRequest field 8) makes AgentService return an empty turn.
- Send ModelDetails (field 3); thinking variants (Composer, Grok, *-thinking)
return an empty turn when only requested_model (field 9) is set.
- Route tool-call history and declared tool schemas through AgentService:
encode OpenAI tools into mcp_tools (field 4), decode McpArgs and emit real
tool_calls with finish_reason tool_calls.
- Map Composer thinking / Grok thinking_delta (field 4) into visible content
instead of dropping the answer with the unsigned reasoning.
- Ack request_context without echoing MCP tools (double-advertise stalls the
HTTP/2 stream) and ack kv_server_message so the run proceeds.
- Reject IDE builtin execs instead of failing the turn, so the model can
continue with MCP tools or a text answer.
- Add google.protobuf.Value / MCP encoders and a FIXED64 branch to
encodeField in cursorProtobuf.js.
RTK now compresses the source-format body before translation for cursor only:
its translator rewrites role:tool into user XML, so the post-translate pass
missed those tool results. Every other provider keeps the post-translate pass
unchanged.
Replace the retired api-inference.huggingface.co host with the Inference
Providers router (router.huggingface.co): imageConfig.modelMap resolves
Hub ids to provider-resolved ids, image-to-image models receive the
source image in inputs with the prompt under parameters.prompt, and a
new sttConfig wires the hf-inference ASR route. The image catalog grows
to 23 models, dead whisper-small is replaced by whisper-large-v3-turbo,
the unusable "language" param is dropped, and edit models declare the
edit capability so the dashboard offers a source image. Adds unit and
end-to-end coverage plus a model-id guard on custom endpoints.
Free-tier Zen models reject Responses requests with 403 FreeTierError
when client tools are present but the fingerprint quartet is missing.
Apply the fingerprint tools to every OpenCode request, canonicalise
case variants of the quartet (Bash->bash) without duplication, and
restore the caller's original spellings on the response side via a
request-local WeakMap threaded through the existing toolNameMap.
A stream that stalled or lost its upstream was closed with no terminal frame
at all, so clients saw "200 OK, a few chunks, then nothing" and could not tell
a truncated reply from a finished one. The Responses passthrough path already
synthesized response.failed; every other client format got nothing.
The watchdog now hands its reason ("stream stall timeout" or "upstream
connection lost") to onAbortTerminal, and buildStreamErrorBytes frames it per
client format: OpenAI-compatible clients get data: {"error":{...}} followed by
data: [DONE], Anthropic clients get `event: error`. The error frame always
precedes [DONE] (openai-python raises APIError on any data payload carrying an
error key), and no synthetic finish_reason is ever emitted — a truncated
stream must not look like a clean stop.
Co-Authored-By: Claude Code <noreply@anthropic.com>
A Vertex job id is a base64url-encoded operation name, and base64url decoding
accepts arbitrary bytes without throwing, so the previous decode-and-split
check let a crafted id splice a path traversal into the fetch URL while the
Bearer token stayed attached — e.g. "..%2F..%2Fevil" resolved to
/v1/evil:fetchPredictOperation on the Vertex host. body.model had the same
shape on the create path, where it is interpolated into the URL unescaped.
decodeJobId now requires a charset-only id, a byte-for-byte round-trip, and a
decoded name matching ^projects/{p}/locations/{l}/publishers/{pub}/models/{m}/operations/{op}$
— no field may contain "/", so ".." can never reach the URL. model ids are
restricted to [A-Za-z0-9._-].
Co-Authored-By: Claude Code <noreply@anthropic.com>
The merged Qoder work also rewrote shared translator/handler code so that
/v1/responses clients got token usage on response.completed. That changed
behaviour for every provider, not just Qoder: proxies saw input tokens
rise by the 2000-token context buffer, and the plain token mapping was
replaced by one that always adds input_tokens_details.
A probe confirms the Qoder benefit does not depend on those edits: the
executor's coalescer already emits one include_usage-style finish chunk, so
a Claude client receives input_tokens and cache_read_input_tokens with
every shared file at its original state. Only the Responses path relies on
the shared translator, and that path has no Qoder-owned seam to put it in.
Reverts the shared files to their pre-PR state and drops the Responses
usage test. The Cline envelope unwrap in nonStreamingHandler.js, which
landed after the PR in the same file, is kept.
defaultClaudeToolType() stamped tools[].type = "custom" onto every
Claude-format request carrying tools since e08ac6da. That satisfied
MiniMax (error 2013) but broke Anthropic-compatible endpoints that only
accept the legacy typeless tool shape. DeepSeek's endpoint
(api.deepseek.com/anthropic/v1/messages) whitelists its tool `type` enum
to the web_search_* variants and answers HTTP 400 "unknown variant
`custom`", so every Claude Code request routed to a DeepSeek connection
failed and surfaced as a persistent 503.
Run the defaulting only when the target provider declares the new
requireClaudeToolType quirk (MiniMax, MiniMax-CN). Add
shouldDefaultClaudeToolType(provider, finalFormat, tools, PROVIDERS) in
translator/concerns/toolCall.js so the gate is unit-testable, and cover
MiniMax keeping the explicit type, DeepSeek/Anthropic staying typeless,
non-Claude formats and tool-less requests never defaulting.
Effectively a no-op for MiniMax and a restore of the pre-e08ac6da
behaviour everywhere else. Another strict gateway now only needs the same
one-line quirk instead of a global behavioural change.
Cline (api.cline.bot) wraps non-stream chat completions in
{"success":true,"data":{...choices...}}, which both the dashboard model-test
ping and the proxy non-stream path read at top level, producing "Provider
returned no completion choices for this model" (#3644). Unwrap the envelope
before usage extraction and response translation; the error envelope
({"success":false,...}) never matches and passes through untouched.
Scoped through `transport.quirks.clineEnvelope` so only cline/clinepass opt
in — no other provider's response body is ever rewritten.
Also adds a live Cline catalog: `fetchClineRawModels()` is shared between
`resolveClineModels()` (full catalog, including free-tier ids such as
z-ai/glm-5.3-flash) and `resolveClinepassModels()` (cline-pass/* only), wired
into /v1/models, the per-provider models route, and the combo selector's
model picker with the static catalog kept as fallback.
Refreshes the dead api-airforce free models (anthropic/claude-3.7-sonnet,
moonshot/kimi-k2.6, google/gemini-2.5-flash) with the live gpt-oss-120b,
gpt-oss-20b and kimi-k2.7-code, plus passthroughModels, forceStream and a
suggested-models filter.
- Coalesce Qoder's empty finish-in-delta frame with the later choices:[] usage
frame so OpenAI and Claude clients receive prompt_tokens, completion_tokens
and cache-hit tokens (the dashboard already saw them)
- Upload inlined images through /api/v2/image/upload like qodercli, and stub
oversized non-image files instead of stuffing 30MB+ data URIs into
agent_chat_generation
- Emit response.completed -> response.usage for chat-native upstreams so
/v1/responses clients (Codex CLI, sub2api) no longer log 0/0/0
- Keep Claude message_delta.usage working when usage arrives without choices[0]
- Escalate to the smallest advertised Qoder context tier (200K/400K/1M) when
the estimated prompt no longer fits max_input_tokens
- Pass apiKey for PAT connections and list hidden enable:false catalog keys
from /v1/models
- Add gpt-image-1.5, gpt-image-2, gpt-image-2.5, gpt-image-2.5-flare and
gpt-image-2.5-sunburst as Codex image models with multi-image support
- Add gpt-image-2.5, gpt-image-2.5-flare and gpt-image-2.5-sunburst to the
OpenAI provider catalog
- Route tool-backed image models through the Codex responses model while
passing the selected model to the image_generation tool, pinning
tool_choice and deriving generate/edit from the presence of references
- Cover the Codex gpt-image-2.5 request shape with a unit test
Video generation was xAI-only. Adds an adapter layer under
open-sse/handlers/videoProviders/ so /v1/videos/* can target OpenRouter or
Google Cloud credentials. A provider with no adapter keeps the exact previous
behaviour (raw body to {baseUrl}/{action}, poll {baseUrl}/{id}, verbatim
passthrough), so the xAI path is unchanged.
- openrouter: async job shape identical to xAI; creation POSTs to the /videos
collection root (no /generations suffix) and the registry HTTP-Referer /
X-Title headers are applied. Bodies pass through verbatim.
- vertex: two-way translation, since Veo does not speak the OpenAI-ish videos
shape. create -> :predictLongRunning { instances[], parameters{} }, poll ->
:fetchPredictOperation (Veo has no REST GET poll). The operation resource
name is base64url-encoded into the job id so GET /v1/videos/{id} stays a
flat path. Access tokens are minted from Service Account JSON via the
existing refreshVertexToken; raw API keys are rejected up front. The
operation response maps back onto the { id, status, video, videos } shape
clients already poll.
- videoCore: the request plan is rebuilt per attempt, so the 401 -> refresh
once -> retry once path picks up the refreshed token. Adapter validation
errors return 400 before any upstream call, so a malformed request can never
create a billable job.
- videoGeneration: GET /v1/videos/{id} resolves the provider from the pinned
x-connection-id connection, then ?provider=, then falls back to the xAI
default.
- registry: openrouter and vertex gain videoConfig, the video serviceKind and
video-kind models (Veo 3.1 / 3 / 2, Sora 2 Pro, Seedance 2.0).
The image handler's `version` header was commented out, so Codex image
requests reached chatgpt.com without the Version identity the backend
expects. Restore it and route every Codex identity header through one
constant.
The CLI version now lives on registry codex.transport as `cliVersion`
(the same pattern gemini-cli uses) and is re-exported as CODEX_CLI_VERSION,
so the registry User-Agent, the image handler and the connection test can
no longer drift apart. Bumped 0.136.0 -> 0.154.0 (current stable).
Co-Authored-By: Claude Code <noreply@anthropic.com>
- add a dedicated OpenCode Go executor that always sends x-opencode-session
- preserve a valid caller-provided native OpenCode session header
- translate downstream Agent session IDs into opaque, stable, Agent-scoped IDs
- forward the original provider session seed and client tool on both initial and credential-refresh requests
Non-streaming codex traffic recorded cached_tokens: 0 even when upstream
prompt caching worked. The Claude-format branch (which OpenAI Responses
usage also matches) never read input_tokens_details, and the OpenAI
branch ignored a top-level flat cached_tokens. Read both in both
branches; Responses prompts are cache-inclusive so canonicalizeUsage
passes the value through without folding. 5 new regression tests.
Closes four SSRF guard bypasses reported in #3714:
- Block alternate IPv6 encodings (hex format, NAT64, IPv4-compatible, IPv4-mapped) by parsing to 16-bit groups
- Normalize trailing dots on hostnames to prevent FQDN bypasses
- Add assertPublicUrlResolved() with DNS resolution to block wildcard DNS domains resolving to private/metadata IPs
- Add fetchPublic() to safely handle and validate HTTP redirects
The 3000 ms timeout on /v1/compress was fixed, so busy or slow machines
timed out often and sent the LLM an inconsistently compressed body,
hurting prompt caching. Add a headroomTimeoutMs setting, thread it from
the chat handler down to compressWithHeadroom, expose it in the Token
Saver dashboard, and normalize invalid values back to the 3000 ms default.
Prefer the sourceFormat-matched runtime transport over a model's
declared targetFormat when both apply. MiniMax-M3 previously resolved
to a Claude-shaped body while being posted to the already-selected
OpenAI endpoint, silently dropping image_url blocks from OpenAI
clients. Fixes#3418.
Resolve body.size through sizeToAspectRatio and append the ratio as a
-WxH suffix so the executor's parseImageConfig picks it up. Also fall
back to gemini-3.1-flash-image when a non-image model reaches the
image handler.
The separate zai-search entry showed "No connections" on the web search
page because credentials live on the `glm` connection, not on it. Every
other provider that does both chat and search (antigravity, kimi, xai,
gemini) declares webSearch on the provider itself, so do the same here.
- glm gains serviceKinds ["llm", "webSearch"] and the MCP searchConfig
- the request builder / normalizer move from "zai-search" to "glm"
- drop the zai-search registry entry and its svg logo, which also
removes the only need for svg logo support in getProviderIconSrc
ollama-search keeps its own entry and credentialFallback: its search
endpoint is unrelated to the ollama chat transport.
Register two web search providers that reuse an existing chat provider's
API key instead of requiring their own connection:
- ollama-search (POST ollama.com/api/web_search) reuses the `ollama` key
- zai-search (POST api.z.ai MCP web_search_prime) reuses the `glm` key
A new `credentialFallback` registry field drives this: when a search
provider has no connection of its own, the search handler falls back to
the linked chat provider's credentials.
Also teach getProviderIconSrc to serve .svg logos for providers that
ship vector art.
Strict Anthropic-compatible gateways (e.g. MiniMax) reject Claude-format
requests with HTTP 400 when tools[].type is missing. Normalize each
missing/falsy tools[].type to "custom" before dispatch when the final
request format is Claude. Built-in tool types (computer_use, bash,
web_search_*) are passed through untouched.
Route POST /v1/search with provider "antigravity" through Google Search
grounding on v1internal:generateContent, using the existing Antigravity
OAuth account pool. Grounding chunks become citations with the grounded
sentence as snippet and its surrounding answer text as content.
Upstream repeats a source across chunks, so citations are keyed by URL
and their snippets merged. A missing projectId is reported up front —
upstream answers a fabricated or absent project with a misleading
"no valid license" 403.
Based on the approach in #3437 by @Nautilaceae.
Xquik needs a GET request with x-api-key auth and a tweets envelope
normalizer, neither of which the generic search fallback provides. Adds a
dedicated request builder and normalizer, cursor pagination passthrough,
result-based credit usage reporting, and a validateUrl probe so key
validation hits the no-charge credits endpoint.
opencode-go hard-coded targetFormat: claude per model, so every client
format was force-routed to /messages (Codex/OpenAI clients paid a lossy
Responses->OpenAI->Claude double translation). Declare the existing
upstream multi-endpoint transports [openai, claude, openai-responses]
and guard per model via registry supportedFormats: kimi/glm/mimo only
support /chat/completions, minimax/qwen add /messages, deepseek adds
/responses. Undeclared models keep the upstream default.
Drop the bespoke OpenCodeGoExecutor (its shared _lastModel cache could
cross auth headers between concurrent requests); DefaultExecutor already
consumes runtimeTransport and injects reasoning content.
Antigravity and gemini-cli wrap their payload in { response: {...} }.
extractUsageFromResponse only tested top-level usageMetadata, so every
non-streaming antigravity request logged zero usage (IN 0 | OUT 0) and
zeroed rows in the usage dashboard. Read the envelope the same way
usageTracking.js and nonStreamingHandler.js already do; top-level
metadata keeps priority and the OpenAI/Claude branches are untouched.
Fixes#3260
Registry entry plus one config-driven FORMAT_HANDLERS handler. The model id
travels in an HTTP `model` header rather than the JSON body, and the voice is
a reference_id (preset or cloned voice model).
Closes#2411
Passthrough kept the client's own cache_control markers, which point at
pre-normalization offsets. Once normalize/dedupe reshaped system and tools,
the breakpoints landed mid-array and the tail was re-cached every request.
- Pin the last system block and last tool at ttl 1h (was the client's 5m)
- Anchor the last assistant turn at 5m, falling back to the final message
so a first turn still gets a breakpoint
- Fold mid-conversation system messages into the neighbouring user turn
instead of hoisting them into body.system, where the volatile token
counters invalidated the prefix on every request
- Run the anchoring after every token saver, at the final body
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Pool/IP fitness registry (globalThis-backed, provider::model scopes, 5-min
cooldown, provider::* wildcard) fed by pool-scoped failures: freebuff
limited-IP / model-locked gates and opencode free per-IP limits. chatCore
retries a failed pool via another pool without locking the account; new
Smart rotation strategy (per-connection + no-auth providers) skips unfit
pools. Proxy Fitness dashboard page: active-block table with provider/IP
filters, per-record Clear and provider-scoped Clear All; egress column via
pool geo enrichment. Unit tests for the registry.
Add Self-hosted STT/TTS/Embedding providers that read baseUrl per connection
instead of a fixed registry endpoint, so 9Router can point at whisper.cpp,
faster-whisper, Kokoro-FastAPI, llama-server, vLLM, Infinity, and similar
OpenAI-compatible local servers.
Self-hosted Embedding refuses to run without a baseUrl rather than falling
back to api.openai.com like openaiCompatNode does, since that fallback would
silently send input text and the API key to OpenAI under a provider named
"Self-hosted". Also fixes embeddingsCore to catch adapter build errors as a
400 instead of letting them escape uncaught, and bounds the upstream fetch
with FETCH_CONNECT_TIMEOUT_MS to avoid hanging forever on a dead endpoint.
Self-hosted TTS treats a bare model value as the model rather than the voice,
since the generic OpenAI TTS convention (bare = voice) is backwards for a
provider where the model is the variable part.
Codex Responses Lite clients routed to a chat-native OpenAI-compatible
provider lost tool use in three places: non-streaming Chat responses
leaked the raw chat.completion envelope instead of Responses output
items, internal reasoning continuity fields leaked into the outbound
Chat body causing some upstreams to reject the request, and the
Responses to Chat request translator ignored additional_tools,
custom_tool_call, and custom_tool_call_output items entirely.
Also fixes apiType (chat vs responses) for openai-compatible nodes
being resolved from the immutable provider ID instead of the stored
node config, so editing a node's API Type had no runtime effect.
Adds mimo-v2.5-tts as a Media Provider TTS through the existing
OpenAI-compatible chat-completions endpoint. Voice is selected via the
top-level audio.voice field, and an optional style/language hint is
threaded through tts.js -> ttsCore.js -> the new adapter.
Add "ultra" reasoning level for Codex GPT-5.6 Sol and Terra, and expose
Max for Luna (Luna falls back Ultra to Max since it is not supported
upstream). Scoped to cx/ routes only; Kiro and generic OpenAI routing
unchanged.
handleForcedSSEToJson dropped cached prompt tokens in two ways: the
Responses branch summed only input_tokens, which excludes cache_read
and cache_creation on cache-capable upstreams (measured 2012 reported
vs ~5344 actual, 5332 from cache); and the Chat Completions branch
computed usage correctly but it didn't always reach the client (an
Anthropic response with cache_read_input_tokens: 11022 arrived with no
usage field at all). Now folds cache counters into prompt_tokens,
surfaces them via prompt_tokens_details, and re-attaches usage before
serialisation.
Remove "Ported from OmniRoute" and cockpit-tools attribution comments.
User-Agent strings and README/landing credits are left intact.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Rotating-RT providers (xAI/grok-cli) issue a new refresh_token on every
refresh; mutate credentials in-place so refreshWithRetry reuses the fresh
RT instead of the already-consumed one.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
asepharyana
decoluaandClaude Code
dinhkarate
Amir Seify
kimono381
MUH. IQRAM BAHRING
Aaron
LLL
galiehneh
Nick Nyanjui
Hai Trinh
coozgan
turingcat
Sami Basra
IEatCodeDaily
openhands
snower
fasilu
warelik
kriptoburak
KiMelody
Nguyen Thanh Dat
zmf
omar-nahhas
nguyenha935
MiQieR
seakleang.nhak