mirror of
https://github.com/instructkr/claw-code.git
synced 2026-04-12 11:04:51 +08:00
docs(roadmap): add #63 — droid session completion semantics broken
Documents the late-arriving droid output issue discovered during ultraclaw batch processing. Sessions report completion before file writes are fully flushed to working tree. Source: ultraclaw dogfood 2026-04-12
This commit is contained in:
@@ -491,3 +491,5 @@ Model name prefix now wins unconditionally over env-var presence. Regression tes
|
||||
61. **`OPENAI_BASE_URL` ignored when model name has no recognized prefix** — user report 2026-04-10 in #claw-code (MaxDerVerpeilte, Ollama). User set `OPENAI_BASE_URL=http://127.0.0.1:11434/v1` with model `qwen2.5-coder:7b` but claw asked for Anthropic credentials. `detect_provider_kind()` checks model prefix first, then falls through to env-var presence — but `OPENAI_BASE_URL` was not in the cascade, so unrecognized model names always hit the Anthropic default. **Done at `1ecdb10` 2026-04-10**: `OPENAI_BASE_URL` + `OPENAI_API_KEY` now beats Anthropic env-check. `OPENAI_BASE_URL` alone (no key, e.g. Ollama) is last-resort before Anthropic default. Source: MaxDerVerpeilte in #claw-code; traced by gaebal-gajae.
|
||||
|
||||
62. **Worker state file surface not implemented** — dogfooded 2026-04-10. ROADMAP "Deployment Architecture Gap" section documents `.claw/worker-state.json` as the "canonical observability contract" for clawhip, but `emit_state_file()` is not wired in `WorkerRegistry::transition()`. The file is never written. **Status:** documentation exists, code does not. Fix shape: wire `emit_state_file()` to write `.claw/worker-state.json` atomically on every `WorkerStatus` transition; add `claw state` CLI subcommand to read/display it. Blocker: none. Source: Jobdori dogfood.
|
||||
|
||||
63. **Droid session completion semantics broken: code arrives after "status: completed"** — dogfooded 2026-04-12. Ultraclaw droid sessions (use-droid via acpx) report `session.status: completed` before file writes are fully flushed/synced to the working tree. Discovered +410 lines of "late-arriving" droid output that appeared after I had already assessed 8 sessions as "no code produced." This creates false-negative assessments and duplicate work. **Fix shape:** (a) droid agent should only report completion after explicit file-write confirmation (fsync or existence check); (b) or, claw-code should expose a `pending_writes` status that indicates "agent responded, disk flush pending"; (c) lane orchestrators should poll for file changes for N seconds after completion before final assessment. **Blocker:** none. Source: Jobdori ultraclaw dogfood 2026-04-12.
|
||||
|
||||
Reference in New Issue
Block a user