docs(repo): add repo-wide team-review agent skill - #4004
Merged
Conversation
The four-expert review workflow lived only in a personal ~/.claude command with a hardcoded local dump directory, so nobody else could run it and it never got reviewed alongside the code it reviews. Move it under .claude/skills/ next to the connector skills. The report now lands in the agent session scratchpad rather than a machine-specific path, and the target argument, the expert agent names and the no-findings path are spelled out so a run does not rely on conventions only the original author knew. Skills carry no disable-model-invocation flag, so the description states the skill is user-invoked only: one run spawns roughly ten subagents.
The description asked the agent not to invoke the skill on its own, which is prose the agent may ignore, and the commit that added it wrongly claimed skills have no frontmatter flag for that. Set disable-model-invocation, which drops the description from the agent's context and blocks the Skill call outright. AGENTS.md now says so and asks the agent to suggest /team-review instead of loading it. The target binds through $ARGUMENTS since /team-review is the only entry point left.
Experts sometimes reach for cargo test to settle a claim. Parallel runs from four agents serialize on the target-dir lock and burn the reading budget. The charter now requires a stated reason before any test or build run. The report ends with per-severity counts so the verdict can be read without scanning the sections.
hubcio
force-pushed
the
docs/team-review-skill
branch
from
September 1, 2026 06:12
8e70035 to
3ff7671
Compare
numinnex
approved these changes
Sep 1, 2026
spetz
approved these changes
Sep 1, 2026
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
5 participants
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.
This PR adds a new skill used to review changes.
Contributors are encouraged to use it.
It is placed under .claude/skills/ next to the connector skills.
The report lands in the agent session scratchpad.
Skills carry no disable-model-invocation flag, so the
description states the skill is user-invoked only: one run
spawns roughly ten subagents.