Repository navigation
Standards index: re-pin SLSA v1.2, Scorecard v5.7.0, MCP 2026-07-28, ACS v0.1.4 - #2432
Open
remigiusz-antczak wants to merge 2 commits into
Open
remigiusz-antczak wants to merge 2 commits into
remigiusz-antczak wants to merge 2 commits into
Conversation
…ACS v0.1.4 Re-fetches each drifted source and re-pins it, verified 2026-11-19. SLSA rows move from v1.0 to v1.2 (levels now at build-track-basics, requirements at build-requirements; the levels row quotes the v1.2 consumer requirement). The four Scorecard checks-catalog rows pin the commit tagged v5.7.0; every quoted check text is unchanged. The three MCP rows move to the current 2026-07-28 revision (annotation definition now capitalized in source). The ACS row pins tag v0.1.4: the spec is still v0.1.0, and the README now names ACS-Core and "conformance profiles". security-ai-agents.md cites the same versions and fetch dates. The Anthropic code-review and pr-review-toolkit plugin rows, and the capability matrix link, point to anthropics/claude-plugins-official, where the plugins moved; the parent-repo stats are refreshed for that repository. The OpenCode TUI source row is left unchanged: the moved page no longer documents /compact, /thinking or ctrl+t, so its quotes need a separate re-audit. No-Mechanism-Reason: citation re-pin only (source versions, URLs and fetch dates); no rule or doctrine changed, so there is no eval or script behavior to add, and no existing test checks these pins. Tests: 470/470 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The ACS version note added in this branch grew the reference from 39145
to 39166 bytes, which broke the size ratchet, the "agent / LLM / MCP"
must-load ceiling (18479 vs 18474 tokens), and the Phase 0-2 floor pin.
Reword the surrounding prose ("another pre-stable spec", "not a risk
catalog", "newest tag v0.1.4 as of 2026-11-19") so the file measures
39144 bytes. That is back to the base token count, so the floor pin
holds exactly. The version pins, dates and URLs are unchanged.
Local checks: size-ratchet ok, mustload ok (phase floor LIGHT
18616/18616, FULL 26595/26595), and the three self-tests that failed
in CI now pass. The other 8 failures in local shard 1 come from the
sandbox and host load: serial_gate cannot see processes and
heavy_gate sees host load. They passed in CI on the previous head.
No-Mechanism-Reason: wording trim to keep budgets, no behavior change
Tests: 3/3
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Re-fetched the standards the skill cites and moved four of them to their current versions, so the citations match what the sources say today.
Changes
Not in this PR
Tests
Pre-push
run-ci-tests.sh --affected: all 13 selected suites passed. An independent review found 0 red, 1 yellow and 1 blue, and both are fixed in this commit.🤖 Generated with Claude Code