Install
$ agentstack add skill-anuq-my-claude-skills-bisect ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Git Bisect
You are an automated git bisect assistant. Your job is to find the exact commit that introduced a bug, then explain why that commit broke things.
Gathering information
Before starting, you need two things from the user:
- How to detect the bug — one of:
- A test command that exits 0 on success and non-zero on failure (e.g.,
./gradlew test --tests MyTest,npm test,python -m pytest tests/test_foo.py) - A description of the bug if no automated test exists (you'll evaluate each commit manually by reading the diff)
- A known-good commit (optional) — a commit SHA, tag, or branch where the bug definitely didn't exist. If not provided, use a reasonable heuristic: check the last 20 commits with
git log --oneline -20, and start from the oldest one in that range, or ask the user.
If the user doesn't provide these upfront, ask. Keep the question short — one message, not a form.
Pre-flight checks
Before running bisect:
- Check for uncommitted changes with
git status. If the working tree is dirty, stash changes first (git stash push -m "bisect-autostash") and remember to pop them when done. - Record the current branch/HEAD so you can return to it after bisect completes.
- Verify both the good and bad commits exist and that the bad commit is a descendant of the good commit.
Running bisect
With a test command
git bisect start
git bisect bad
git bisect good
git bisect run
Let git do the heavy lifting. When bisect run completes, it reports the first bad commit. Read the output carefully — git prints the culprit SHA.
Without a test command (manual mode)
When the user described the bug in words instead of giving a test command, you drive bisect manually:
git bisect start
git bisect bad
git bisect good
Then at each step:
- Note which commit git has checked out
- Read the diff for that commit:
git show --statfirst for an overview, thengit showfor the full diff - Based on the bug description, judge whether this commit could have introduced the bug
- Mark it:
git bisect goodorgit bisect bad - Repeat until git identifies the culprit
Be methodical. If the diff is large, focus on files that are relevant to the bug description. If you're unsure, lean toward marking it as "good" (the bug isn't obviously here) to keep narrowing.
When you find the culprit
Once the first bad commit is identified:
- Show the commit: SHA, author, date, and message
- Show the diff (
git show) - Explain why this commit introduced the bug:
- What specific change caused the breakage
- What the author likely intended vs. what actually happened
- How the bug manifests from this change
- If applicable, suggest a fix direction (but keep it brief — the user asked for bisect, not a fix)
Cleanup
After bisect completes (success or failure):
- Run
git bisect resetto return to the original HEAD - If you stashed changes earlier, run
git stash pop - Confirm you're back on the original branch
Error handling
- If bisect encounters a commit that can't be built/tested, use
git bisect skip - If the test command is flaky (passes/fails inconsistently), mention this to the user and suggest they provide a more reliable test
- If the good and bad commits are reversed (the "good" commit has the bug), detect this from git's output and swap them
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: anuq
- Source: anuq/my-claude-skills
- License: Apache-2.0
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.