Skip to content

Protect all 35 verified offline retrieval cases - #147

Merged
ayhammouda merged 1 commit into
mainfrom
agent/63-ratchet-35-main66
Oct 5, 2026
Merged

ayhammouda merged 1 commit into
mainfrom
agent/63-ratchet-35-main66

Conversation

@ayhammouda

@ayhammouda ayhammouda commented Oct 5, 2026 •

Copy link
Copy Markdown
Owner

Refs #63.

User problem and decision

The historical frozen gate protects 8 hits while verified merged main has 35. Protect the verified gains with exactly 27 false→true baseline flags (original 26 plus EX-015). No other baseline bytes change: 65 case keys, citations, timings, corpus SHA, 1,603-document floor and two-second ceiling are preserved. No source/scorer/corpus/tests/workflows changed. This strengthens the gate; it does not improve runtime retrieval or measure answer accuracy.

Owner decision: #63 (comment)
Base: 66bd44d. Local committed head ccfbe48b873bdd440aacfddc8ed2155fe4412cb5; published head 94d9112 has matching tree 0b78e54982d5e2bc1d91a270ede070ac0bf3ed56 and exact base.

Acceptance and evidence

  • Owner inspected only 27 upward flags; reversing tokens reproduces exact base bytes.
  • Fresh isolated developer locked sync, full-path ruff, pyright, 596 tests, corpus validation, doctor and build passed; 4 installed-wheel stdio tests passed.
  • Base/head retained-index replays both 35/65 hits, 69 citations, 1,603 documents; no case outcome/citation differences.
  • All 27 saved negative controls yield only their lost-hit error under the new baseline and no error under old baseline.
  • Broker independent prepublication review approved publication; published head 94d911284b78df4c34e60bc0e4165ef145e85da8 has the exact owner-reviewed tree and base.
  • Detailed fresh pinned-source/27-negative-control verifier evidence will be exposed and checked with exact-head pdctl verify in the next full review window; publication output itself only returned the approved ref.
  • Exact published-head verification, required hosted CI, review triage/resolved threads, SHA-matched merge and post-merge CI.

Supervisor review

Vision explicitly authorizes only upward regression-baseline strengthening. Independent justification is mandatory. No test weakening, corpus answer edits, new behavior or unrelated dependencies.

Limitations and review

Developer index was retained, not rebuilt; those results do not replace independent source-pinned evidence. One existing packaging deprecation warning. No generated-answer, timing-improvement or named-client claim. CodeRabbit skipped this baseline-only change (path filters; another integration reports fewer than 10 stars). No actionable inline/submitted findings or review threads at 2026-10-05 18:37 UTC. A skipped review is not approval; broker independent review remains mandatory. Outcome review 2026-10-17; reject nonreproducible protections or downward relaxation and revisit on actual regressions/feedback.

@coderabbiteu

coderabbiteu Bot commented Oct 5, 2026

Copy link
Copy Markdown

Important

  • 🔍 Trigger review

This repository does not receive automatic reviews because it has fewer than 10 stars.

⚙️ Run configuration
  • Configuration used: Repository: ayhammouda/python-docs-mcp-server/.coderabbit.yaml
  • Review profile: CHILL
  • Plan: Advanced
  • Run ID: 3c2ef714-8751-40fc-9b30-10e99fc6428d
  • Autopilot · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

Comment @coderabbitai help to get the list of available commands.

@coderabbitai

coderabbitai Bot commented Oct 5, 2026

Copy link
Copy Markdown

Important

Review skipped

Review was skipped due to path filters

⛔ Files ignored due to path filters (1)
  • docs/benchmarks/regression-baseline.json is excluded by none and included by none

CodeRabbit blocks several paths by default. You can override this behavior by explicitly including those paths in the path filters. For example, including **/dist/** will override the default block on the dist directory, by removing the pattern from both the lists.

⚙️ Run configuration
  • Configuration used: Repository: ayhammouda/python-docs-mcp-server/.coderabbit.yaml
  • Review profile: CHILL
  • Plan: Advanced
  • Run ID: 62bc15f3-00d8-4295-a306-4e15ed2bf0ef

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

  • 🔍 Trigger review
  • Autopilot · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

Comment @coderabbitai help to get the list of available commands.

@ayhammouda

Copy link
Copy Markdown
Owner Author

Vision — automated project maintainer

Independent verification (temporary credential): success

Head: 94d911284b78df4c34e60bc0e4165ef145e85da8; main: 66bd44d0902d644f7fad80827a3739bd3eda5130.

{
  "head_sha": "94d911284b78df4c34e60bc0e4165ef145e85da8",
  "base_sha": "66bd44d0902d644f7fad80827a3739bd3eda5130",
  "approved": true,
  "summary": "Approved for prepublication review. The full diff changes only 27 regression-baseline flags from false to true; reversing them recovers the exact base file. A fresh source-pinned index produced 1,603 documents. Its offline replay reproduced 35 hits and 69 citations across 65 cases, and all 27 individual negative controls behaved as intended. Locked checks, doctor, package build, and installed-wheel stdio smoke passed. Public main still matches the base; hosted CI for this unpublished head is pending, not a local verification pass.",
  "commands": [
    {
      "command": "git clone --depth=1 https://github.com/ayhammouda/python-docs-mcp-server.git repo",
      "exit_code": 0
    },
    {
      "command": "git fetch --depth=1 origin 94d911284b78df4c34e60bc0e4165ef145e85da8",
      "exit_code": 0
    },
    {
      "command": "git fetch --depth=1 origin 66bd44d0902d644f7fad80827a3739bd3eda5130",
      "exit_code": 0
    },
    {
      "command": "git diff --check 66bd44d0902d644f7fad80827a3739bd3eda5130 HEAD",
      "exit_code": 0
    },
    {
      "command": "/usr/local/bin/uv sync --locked --dev",
      "exit_code": 0
    },
    {
      "command": "/usr/local/bin/uv run --locked ruff check src/ tests/ benchmarks/ ops/ .github/scripts/",
      "exit_code": 0
    },
    {
      "command": "/usr/local/bin/uv run --locked pyright src/ benchmarks/",
      "exit_code": 0
    },
    {
      "command": "/usr/local/bin/uv run --locked pytest --tb=short -q",
      "exit_code": 0
    },
    {
      "command": "/usr/local/bin/uv run --locked python -m benchmarks validate-corpus --corpus docs/benchmarks/corpus.yml --schema docs/benchmarks/corpus.schema.json",
      "exit_code": 0
    },
    {
      "command": "/usr/local/bin/uv build --no-sources",
      "exit_code": 0
    },
    {
      "command": "XDG_CACHE_HOME=\"$PWD/.verifier-cache\" /usr/local/bin/uv run --locked python-docs-mcp-server build-index --versions 3.11,3.12,3.13",
      "exit_code": 0
    },
    {
      "command": "/usr/local/bin/uv run --locked python -m benchmarks.regression --index .verifier-cache/mcp-python-docs/index.db --output .review-actual.json",
      "exit_code": 0
    },
    {
      "command": "/usr/local/bin/uv run --locked python - [independent baseline and 27 negative-control assertions via stdin]",
      "exit_code": 0
    },
    {
      "command": "XDG_CACHE_HOME=\"$PWD/.verifier-cache\" /usr/local/bin/uv run --locked python-docs-mcp-server doctor",
      "exit_code": 0
    },
    {
      "command": "../.review-wheel-venv/bin/python -m pytest --import-mode=importlib test_stdio_smoke.py -q",
      "exit_code": 0
    },
    {
      "command": "git ls-remote origin refs/heads/main",
      "exit_code": 0
    }
  ],
  "blockers": []
}

@ayhammouda

Copy link
Copy Markdown
Owner Author

Vision — automated project maintainer

Merge decision for exact head 94d911284b78df4c34e60bc0e4165ef145e85da8 against main 66bd44d0902d644f7fad80827a3739bd3eda5130: approve. Broker exact-head independent verification succeeded (#147 (comment)), including a fresh source-pinned 1,603-document replay: 35/65 retrieval hits, 69 citations, and all 27 negative controls. Diff is only 27 false-to-true baseline flags; reversing them restores the old bytes. Locked checks, doctor, build and installed-wheel stdio passed. All 11 hosted exact-head checks pass, current main is unchanged, and no review threads or findings are open. CodeRabbit skipped; it is not the independent approval.

Proceed through SHA-matched pdctl merge, then confirm main CI. This strengthens regression protection only; no runtime or answer-quality improvement and no release claim.

@ayhammouda
ayhammouda merged commit 44c80f9 into main Oct 5, 2026
12 checks passed
@ayhammouda

Copy link
Copy Markdown
Owner Author

Vision — automated project maintainer

Post-merge follow-up for 44c80f91aea9e133c42f81df2ed83c3c53448faf: Product quality (including frozen regression and installed-wheel smoke), CodeQL, Security Audit and four Python 3.12/3.13 matrix jobs passed. Full post-merge CI is not confirmed. Both Python 3.14 jobs and Scorecard were cancelled before any step ran; each annotation says: “The job was not acquired by Runner of type hosted even after multiple attempts.”

Runs: CI, Scorecard. GitHub reports an active hosted-runner assignment incident. This is evidence of infrastructure acquisition failure, not failed product tests.

I attempted the failed-job rerun through the authorized fixed-repository broker; it rejected the endpoint as outside its allowlist. No alternate identity, workflow change, empty commit, check suppression or repeated unsupported request was used. Recovery is pending supported runner reruns; retain all gates and do not claim post-merge completion or a release.

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.

1 participant