Phase 2, OB-3. Every finished turn is checked against fixed rules (the
shared signals.yaml, 19 checks) and the result is stored with the turn.
The Steps panel gets an Issues tab, and the sidebar an amber ring for
warnings not yet looked at.
- The engine records the recording format 1.2 fields the checks read:
the arguments' hash (RFC 8785), repeat_ok, result_error, error_kind,
reaches_network and the todo plan on tool steps; max_iterations on the
turn; max_output on model steps; inbox_at on waits; whether the
machine expects a sandbox in the stamp.
- coworker/observability/signals.py holds the checks, the same ones that
were validated on the in-app recordings and the shared golden test.
- The checks run when a turn ends; a result made by an older
signals.yaml is worked out again on read. New endpoints serve a
turn's issues and mark a session's issues seen.
- The Issues tab groups issues by check, warnings first, with the step's
command and output line and Go to step; "Checked and fine" is folded.
- tests: the Phase 2 golden test (38 scenarios and hash vectors), the
1.2 fields from a real turn, and the endpoints through the server.
A per-model table says whether a model has a thinking switch and which effort levels;
the composer shows a grey effort slider and a brain switch only when the model has them.
llama.cpp and vLLM get thinking as enable_thinking; the provider JSON tables ship in the package.
The server, connectors tool, acceptor and team board no longer install as commands.
They stay runnable as modules (python -m coworker.server.run, …) for development.
A headless one-shot command beside `up`: give it a task, a folder and a model; it builds the same engine the desktop app uses, runs to the end with attendance `auto` (coworker/unattended.py), and leaves a record: every event, the final conversation, a summary with tokens and cost, every answer the engine gave on the absent user's behalf, the audit log, a per-call model log, and a trajectory in the ATIF interchange format. The trajectory and the exit code are the stable contract; the other files are internal.
State and scratch folders are pointed under --out before the engine starts, so a run never touches the machine's own conversations, keys or settings. A transient provider failure or an empty reply is waited out with backoff and a nudge instead of ending the run; permanent errors are not retried. Tool results are bounded with the full text spilled beside the record. A small price table gives an estimated cost per call as a cross-check; an unpriced model gets counts but no cost.
The mode axis decides approvals (`dangerously-bypass-approvals` prints its warning line here); the runner adds no policy of its own.
Linux x86_64 and aarch64 tar.gz from build_linux.sh, attached with versioned and stable names.
The macOS build installs the openshell extra (build_dmg.sh refuses to build without grpcio).
Folders are bind-mounted at their real paths; the runner is a read-only mount and the main command.
The policy is rendered from the session's roots, with the owner's uid and Landlock as a hard requirement.
The runner's pipe is the gateway's gRPC exec stream; the registry caps and cleans up sandboxes.
aisuite comes from PyPI (0.2.0), so the package has no git-URL dependency.
Older top-level spellings (openworker join, up, status) keep working, unlisted.
A machine joins with its own identity and holds a reverse channel to the controller.
Provider keys and connector accounts deploy sealed to the machine's pinned key.
The test suite is broad (79 files) but has no coverage measurement or
threshold, so regressions in under-tested paths go unnoticed.
Add pytest-cov to dev dependencies, configure coverage in pyproject.toml
(branch coverage, source=coworker, show_missing), and update CI to run
pytest with --cov and upload the coverage.xml artifact.
The fail_under floor is set to 0 initially so CI passes without changes
to existing coverage. Once a baseline is established, the floor can be
ratcheted up over time.
Fixes#288
ocw board/journal verbs with server auto-discovery (remote-first; direct SQLite only headless).
ocw board mcp serves the role-scoped toolset to external harnesses; 21 tests incl. both dialects over the real app.
requires-python declares >=3.10 and the README says "Python 3.10+", but
coworker/config.py imports tomllib at module top and tomllib only landed in
the 3.11 stdlib — on 3.10 the package cannot even be imported (every test
module fails collection through the coworker.config import chain), so the
advertised floor is broken in practice.
Fall back to the tomli package (the pre-stdlib implementation of the same API)
and declare it as a dependency only for python_version < '3.11', so 3.11+
installs are unchanged.
Verified on 3.10: importing coworker.config and load_config() parsing a
workspace config.toml both work through the fallback; the config suite still
passes on 3.12.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Imported from andrewyng/aisuite@1b4bbf303e
(contents of its platform/ directory, hoisted to the repo root).
Development history prior to this commit lives in that repository.
Co-authored-by: Devika <devikaverma11@gmail.com>