- unify hook base-env assembly on hooks::base_envs_parts (drop the
hand-copied hook_base_envs in function/agents.rs; graph executor and
spawn-result hooks now share the single construction site, test-pinned)
- factor the duplicated job.started dispatch in jobs.rs handle_start
into one fire_job_started helper
- COYOTE_LLM_ERROR now carries the immediate error (to_string()),
matching COYOTE_TOOL_ERROR; test pins the new format
- remove all stale #[allow(dead_code)] markers in src/hooks.rs
- test infra: rename install_warn_collector -> install_log_collector,
add poison-symmetric warn_snapshot/debug_snapshot helpers, hoist a
shared EnvVarGuard (TestConfigDirGuard now builds on it), add a
same-fire payload-uniqueness test, and give global_tools env
overrides the same malformed-JSON debug diagnostic as global_hooks
- inspection-only flags (--info et al) bypass the first-run wizard:
missing config + info_flag returns Config::default() without writing
anything, in both the sandbox and interactive arms (tests included)
- bound the shim-relocation test's exec in an ETXTBSY retry (10x50ms)
- document the intended 'roles' reserved-name asymmetry at
resolve_bundle_name's gate
Wire the hook engine into the three runtime seams:
- tool.*: ToolCall::eval fires tool.started once the name and arguments
are resolved, then exactly one of tool.completed/tool.failed,
discriminated on both the Err path and tool_call_error-shaped outputs.
The arguments JSON is handed to the engine as the payload; eval itself
performs no payload-file I/O.
- llm.request.*: the three transport free functions
(call_chat_completions, call_chat_completions_streaming, and the quiet
graph-node variant) fire started/completed/failed for every provider
and caller, classifying failures as aborted, api_error (with
COYOTE_LLM_STATUS), or other.
- job.*: handle_job_tool fires job.started before either spawn and the
terminal event inside the detached closures via a pre-resolved
fire_resolved snapshot, so job__start's return is never delayed.
The test sink now captures the payload and cwd of each dispatch
(sink-only change), and seam tests pin event cadence, env contracts, and
byte-identical tool results and errors.
Review findings: the streaming arm now mirrors the non-streaming twin
(openai_extract_responses): partial text/tool calls are returned rather
than discarded by a hard error; only an empty truncated response bails.
Billed usage on truncated turns is recorded before either outcome.
Also: CHANGELOG notes the streaming-incomplete surfacing and corrects
the include/strict phrasing (sent on every responses-wire body).
Parse Responses-API token usage on both wires: non-streaming responses
fill ChatCompletionsOutput.usage and response.completed events feed the
SseHandler accumulator, with cached input mapped from
input_tokens_details.cached_tokens. Harden failure paths: bail with the
payload message on response.failed and error stream events, and surface
incomplete_details.reason when a non-streaming response comes back
incomplete and empty.
Replay hardening: id-only Reasoning blocks (no encrypted_content) are
dropped at replay time since the API 400s on them under store:false,
which would permanently wedge a persisted session; a null summary is
normalized to [] at the same site. Claude's replay guards flip to a
positive Thinking/RedactedThinking allow-list.
Flip the default wire for stock openai (api.openai.com, no custom
api_base) to Responses. The resolver call site now passes a genuine
stock-openai predicate: api-key clients or stock oauth without api_base
qualify, config-oauth gateways never do. Custom api_base and
openai-compatible clients keep defaulting to chat; explicit wire_api
still always wins. Also reject a non-map responses section in
RequestPatch via test.
- assert old Thinking/RedactedThinking blocks re-serialize to the
original wire shape, proving both legs of old-session compat
- add serde_yaml round-trip test for Vec<ThinkingBlock> covering the
on-disk session format, including a Reasoning block with summary
payload and the encrypted_content: None omitted-field case
Add ThinkingBlock::Reasoning to carry OpenAI Responses reasoning items
verbatim. The Responses body builder now requests
reasoning.encrypted_content, marks tool defs strict:false, and replays
each round's reasoning items ahead of the assistant text and function
calls. Both extraction paths (non-streaming and streaming) capture
reasoning output items. Claude and Bedrock replay paths drop the
OpenAI-only variant so mid-session model switches keep working.
Introduce a WireApi enum (responses|chat) with an optional wire_api
config field on the openai and openai-compatible clients, resolved
through a single resolve_wire_api fn: codex forces the responses wire
(an explicit wire_api: chat there is a hard config error), an explicit
wire_api wins everywhere else, and the default stays chat for now.
The openai dispatch generalizes from (RequestData, bool) to
(RequestData, WireApi), and the openai-compatible client gains opt-in
Responses dispatch (URL, body, extractor, and streaming forks).
Both patch layers are now wire-aware: RequestData carries its wire,
RequestPatch gains a responses section, and patch_request_data selects
patches per wire. This deliberately FIXES a codex bug: chat-shaped
model patches (e.g. o3-high's max_tokens/temperature/top_p/
reasoning_effort tweaks) previously merged wire-blind into codex
responses bodies, poisoning them with chat-only keys. On the responses
wire only an explicit responses sub-patch applies; chat-wire patching
is byte-identical to before (pinned by a regression test).
- ClaudeConfig gains an oauth: block (custom endpoints, e.g. a private
gateway); without it, auth: oauth keeps the stock Claude Pro/Max provider
- Claude-Code system-prefix spoof + anthropic-beta header are only sent on
the stock provider path
- OAuth tokens are stamped with an issuer fingerprint (token_url|client_id)
so switching endpoints never replays a token against the wrong issuer
- All auth flows refresh the in-memory access-token cache on save, so
.authenticate mid-session takes effect immediately
- A custom oauth: block now requires api_base: refusing to silently fall
back to api.anthropic.com prevents sending a gateway token to the wrong
host
Graph llm nodes (and the structured-output extractor they share with agent
nodes) now obtain the model reply through a quiet streaming transport,
`call_chat_completions_streaming_quiet`, instead of the non-streaming call.
Because chunks arrive as they are generated, reqwest's client-level
`read_timeout` (default 300s) bounds only a genuine stall between chunks and
the node's own `timeout` becomes the effective bound on total generation.
Previously zero bytes arrived until the whole reply finished, so a 300s
read_timeout silently capped every llm node regardless of its `timeout`
(seen live: code-reviewer's `synthesize` node dying on a 61-file review).
The quiet transport renders nothing (silent SseHandler, no spinner), keeps
tool-call handling identical to the non-streaming path (call-loop detection
off), falls back to the non-streaming request for catalog `no_stream` models
(raced against the abort), and never returns partial output as success.
Supporting changes:
- `sse_stream`/`json_stream` keep the typed `reqwest::Error` in the anyhow
chain so `is_transient_error` still classifies read_timeout stalls as
transient and `max_attempts` retries them (previously stringified).
- `structured::extract` takes the graph/session AbortSignal and refuses to
issue an extraction request for an already-aborted graph.
- `run_chat_loop` gains an injectable CompletionRunner seam (owning client
construction + the before-hook) so the timeout/retry/abort behaviour is
unit-testable over a paused clock; a fake `impl Client` drives the real
trait-default streaming for 600s of virtual time; real-socket tests pin
that read_timeout fires on a stall but not on a slow continuous stream.
Unknown chat/reranker models resolved through the create-from-name
fallback now inherit their provider's family rules (require_max_tokens,
reasoning defaults, etc.) from the quirks overlay, and the model
completion table marks reasoning-capable models with a brain glyph.
Add a hidden --generate-models flag that rebuilds models.yaml from
models.dev and OpenRouter merged with the hand-owned quirks overlay,
regenerate the committed catalog with it (393 -> 1090 models), and
retire the Python updater. GitHub becomes fully quirks-owned since
upstream only tracks Copilot ids, legacy ids upstream dropped are
preserved as hand-owned quirks entries, and non-chat model families
(image/tts/audio/realtime) are excluded from generation.
The Client trait's default chat_completions, chat_completions_streaming,
and embeddings methods now classify failures via ApiStatusError: on a
401 with a cached OAuth token, the token is distrusted (identity-aware
marker) and the call retried exactly once — the retry's prepare step
sees the marker and force-refreshes. Streaming retries only while the
SSE handler has received no content, preventing duplicate rendering.
A second 401 propagates the original error; other retry errors
propagate as-is. API-key clients never retry. No backoff by design:
cost is bounded to one refresh + one retry per failing request.
distrust_access_token compare-and-invalidates the in-memory entry only
when the cached token equals the rejected one, so a concurrent refresh
is never clobbered. is_valid_access_token and both expiry checks in
prepare_oauth_access_token treat marked tokens as expired, forcing a
refresh of provider-rejected tokens that are still locally unexpired.
The marker is cleared after every completed refresh, including ones
that return the same token.
catch_error and sse_stream now bail with ApiStatusError{status, message}
instead of bare anyhow strings, preserving every existing Display output
byte-for-byte. Enables structural status classification (e.g. 401
detection) via downcast through anyhow context chains.
Dark-Alex-17