Skip to content

docs(ai): canbench results are regenerated manually, not by automation#13511

Open
AntonioVentilii wants to merge 2 commits into
mainfrom
claude/noisy-wallet-memory-plan-so0iz0-canbench-docs
Open

docs(ai): canbench results are regenerated manually, not by automation#13511
AntonioVentilii wants to merge 2 commits into
mainfrom
claude/noisy-wallet-memory-plan-so0iz0-canbench-docs

Conversation

@AntonioVentilii

Copy link
Copy Markdown
Collaborator

Motivation

An audit of the benchmark/performance tooling found a documentation inconsistency: docs/ai/governance.md lists canbench_results.yml among "tracked artifacts updated by automation", but no workflow anywhere in .github/workflows/ runs canbench. The file only changes when a developer runs npm run benchmark and commits the result — which is what docs/ai/backend/testing.md ("active but lightly maintained — not gated by CI yet") and docs/ai/pr-and-ci.md correctly describe. The governance table is the entry point agents use to decide whether a file self-updates, so the wrong claim can make them (and humans) assume benchmark drift is handled when it isn't.

Changes

  • Split canbench_results.yml out of the "updated by automation" row in docs/ai/governance.md into its own row stating it is manually regenerated via npm run benchmark, with a link to the canbench section of backend/testing.md.

Tests

Docs-only change; prettier --check passes. (Separately worth considering, not in this PR: actually gating canbench in CI, since results currently drift silently between manual runs.)

🤖 Generated with Claude Code

https://claude.ai/code/session_01GiZ8mmts9aZCCANKQfQ1Xi


Generated by Claude Code

docs/ai/governance.md listed canbench_results.yml among 'tracked artifacts
updated by automation', but no workflow runs canbench — the file is only
updated when a developer runs npm run benchmark and commits the result, as
docs/ai/backend/testing.md correctly states. Split it into its own row with
the accurate regeneration instructions so agents don't wait for an
automation that doesn't exist.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01GiZ8mmts9aZCCANKQfQ1Xi
@AntonioVentilii AntonioVentilii marked this pull request as ready for review July 11, 2026 08:26
@AntonioVentilii AntonioVentilii requested a review from a team as a code owner July 11, 2026 08:26
Copilot AI review requested due to automatic review settings July 11, 2026 08:26
@zeropath-ai

zeropath-ai Bot commented Jul 11, 2026

Copy link
Copy Markdown

No security or compliance issues detected. Reviewed everything up to 52482fc.

Security Overview
Detected Code Changes
Change Type Relevant files
Other relevant categories ► docs/ai/governance.md
    Update governance documentation to adjust listed artifacts (remove and add entries as shown)

Copilot AI left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

Updates the AI governance documentation to accurately reflect how canbench_results.yml is maintained, aligning the “protected paths” table with the repo’s actual benchmark workflow (manual regeneration via npm run benchmark, not CI automation).

Changes:

  • Removes canbench_results.yml from the “Tracked artifacts updated by automation” entry.
  • Adds a dedicated table row clarifying canbench_results.yml is manually regenerated via npm run benchmark.
  • Links the governance entry to the canonical canbench documentation section in docs/ai/backend/testing.md.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants