docs: document CI parity checks - #2668
Conversation
There was a problem hiding this comment.
Graphify reviewed this change.
Looks safe to merge — no coupling regressions and no blocking issues, checked against the code graph (not a self-assessment).
Graphify review — findings
This PR adds a new "CI parity checks" section to the README, documenting the commands contributors can run locally to mirror CI verification (uv sync, pytest, various skillgen checks, and graphify commands). It also notes which tools (Ruff, Pyright, Bandit, pip-audit) are advisory versus blocking. The change is documentation-only within the README, inserted before the existing macOS note and Git workflow sections.
No blocking issues surfaced.
Analysis details — impact, health, verification
Impact & health
Graphify review
Impact — 26 functions depend on the 26 functions this change touches.
Health — grade A; no new coupling hotspots.
Verification — 26 functions in the blast radius were not formally verified this run (proofs are advisory here).
Gate & verification
graphify gate
PASS — objectively clean (no health regressions, tests not run — proofs not run this pass (advisory)). Grounded, not self-assessed.
Advisory (not blocking):
- verification_scope: 26 function(s) in the blast radius were not formally verified this run
|
Shipped in v0.9.42 ( |
Summary
Verification
uv run --frozen ruff check .— passedtools.skillgenconsistency guards — passedgit diff --check— passedScope
Docs-only:
README.md. No source, test, workflow, AGENTS, generated artifact, hook, or skill changes.The full pytest suite was not used as a native-Windows blocker because the repository's Ubuntu CI matrix is authoritative and known Windows parity failures remain in symlink, long-path, POSIX, separator, and encoding behavior.