Repository navigation
docs(harness-ops): refresh four native-overlap dismissals and dismiss brief against Claude Code 2.1.314 - #11433
Conversation
… brief against Claude Code 2.1.314 Re-dismiss rename, resume, stop and agents against their docs-hygiene and session-flow components with the existing reasons, refreshing the component fingerprints that resurfaced them, and dismiss brief against review:brief-reviewer as a shared word only. Rulings follow the owner's recorded direction on #9105; the view is regenerated. Refs: #9105 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01KeBVgbVwsJiQ4E7x6CzZgh
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
|
claude-security-review reviewed No security issues found. The PR changes two data files:
critical-findings: 0 |
|
claude-review reviewed No findings. I read the full diff (
|
Refs: #9105
Summary
This lands the host-run part of the recorded decision on #9105: the dismissals that need fingerprints from a live Claude Code inventory. The inventory came from the installed Claude Code 2.1.314.
Fix
rename/docs-hygiene:rename-references,resume/session-flow:continue-in-background,stop/session-flow:clean-stopandagents/docs-hygiene:write-for-agentswithoverlap.py dismiss, keeping each existing reason word for word. Detect had flagged each oneresurfaced: description changedon the component side only. The native fingerprints were the same on 2.1.314, so the native registration text has not changed and the old reasons still hold.brief/review:brief-reviewer(agent) as a shared word only:/brieftoggles brief-only output mode, and the agent runs the review brief it is dispatched with.docs/native-surfaces.md.Verification
overlap.py detecton the 2.1.314 inventory: the five pairs now appear underdiscovery.suppressed. Exit 3 (degraded) comes from inventory lanes that were already broken (plugin_backed,builtin_tools,builtin_plugins).run-skill-generator: its best pair istesting:run-e2eat 0.2625, below the 0.6 bar, so under decision 5 it gets no entry.overlap.py generate --checkis in sync.overlap.py self-checkexits 3 (degraded) on two advisories that were already there (older recorded extraction versions, no--upstream-sha).scripts/check-changelog-fragments.sh --checkpasses.scripts/check-affected.sh --testson the committed head: 11 passed, 5 skipped.claude-code-docs/discovery:read-docscomplementary/routerow (see Remaining).Remaining
claude-code-docs/discovery:read-docsstore row with verdictcomplementaryand integrationroute, plus the## Boundarysection and description phrase inplugins/discovery/skills/read-docs/SKILL.md. Evidence for it from the 2.1.314 extraction: the surface is abundled-skill,invocable_byismodel+user, and the discovery score is 0.6736.Related
🤖 Generated with Claude Code
https://claude.ai/code/session_01KeBVgbVwsJiQ4E7x6CzZgh