Menu
Git bisect to find the first bad commit by running a test command.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Basé sur la classification professionnelle SOC
Show pull requests that need review, ranked by ease of review. Displays PRs as clickable links with line counts, grouped as a dependency tree when PRs build on each other.
Interactive rebase onto a base branch, resolving conflicts along the way and verifying tests pass. Compares against remote when done.
Evaluate unresolved review comments on the current branch's PR and, after user confirmation, address each in a separate sub-agent and separate commit.
Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Like grill-me, but asks one structured, numbered question at a time with multiple-choice options. Use when the user wants a disciplined interview to stress-test a plan or design, or mentions "interview me".
Evaluate pending (unsubmitted) review comments on the current branch's PR and, after user confirmation, address each in a separate sub-agent and separate commit.
Run bin/claude-review --print and automatically fix all reported issues, committing each fix individually.
| name | bisect |
| description | Git bisect to find the first bad commit by running a test command. |
| user-invocable-only | true |
| allowed-tools | Bash, Read, Glob, Grep, AskUserQuestion |
| argument-hint | <instructions on what command to run to test each commit> |
Use git bisect to binary-search through commit history and find the first commit that introduced a failure.
Check that there are no staged or unstaged changes. If there are, stop and tell the user they need a clean working tree before bisecting, since git bisect checks out older commits.
git status --porcelain
If the output is non-empty, stop and ask the user to commit or stash their changes first.
Use $ARGUMENTS to understand what the user wants to test. This should describe a command or check that fails on the current commit. Determine the exact shell command to run.
Run the test command on the current commit (HEAD). If it passes (exits 0), stop and tell the user — the current commit doesn't exhibit the failure, so there's nothing to bisect.
Start searching backwards from HEAD in steps of 10 commits to find a commit where the test passes.
git stash --include-untracked -q 2>/dev/null; true
git checkout HEAD~10 --quiet
Run the test command. If it still fails, go back another 10 commits (HEAD~10 from the current position). Repeat until either:
Once a good commit is found, note its SHA.
Return to the original branch first, then start bisecting:
git checkout - --quiet
git bisect start
git bisect bad
git bisect good <good-commit-sha>
Git will check out a middle commit. Run the test command on it:
git bisect badgit bisect goodRepeat until git reports the first bad commit.
Save the first bad commit SHA and message. Then reset:
git bisect reset
Tell the user the first bad commit, including:
git show --stat of the commit so they can see which files changed