Install
$ agentstack add skill-thixpin-claude-config-bug-fix ✓ 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
Bug Fix
Fix bugs regression-first: prove the bug exists, fix the cause minimally, prove it cannot return.
Scope
Use for any defect with observably wrong behavior and a known cause — crashes, incorrect output, regressions.
This skill begins once the root cause is identified. If the cause is still unknown, use debugging first, then return here with it.
Do not use for:
- Writing or restructuring tests against working code — use
testing. - Improving healthy code that has no defect — use
code-quality-review.
A vulnerability is a bug: use this workflow and the security-audit checklist together.
Workflow
- Reproduce first. Confirm the bug with a concrete failing case — a command, request, or input that demonstrably misbehaves. If you cannot reproduce it, use
debuggingto establish a reliable trigger before continuing here. - Confirm the root cause. State it in one sentence before fixing, and check the fix targets it rather than the symptom (no null check where the real problem is that the value should never be null). If tracing from symptom to cause is still open work, finish that in
debuggingfirst. - Write the regression test before the fix whenever the project has a test suite and the bug is testable. The test must fail on the current code for the same reason the bug occurs. If an automated test is infeasible (environment-dependent, UI-only, timing), state why and describe the manual verification used instead.
- Fix minimally. The smallest change that removes the root cause. No drive-by refactoring, renames, or style changes in the same edit — propose those separately if worthwhile.
- Verify. Run the regression test, then the smallest relevant surrounding test suite to catch collateral damage.
- Check for siblings. Search for other call sites or copies of the same pattern that share the defect. Fix or report them — a bug that exists in three places and is fixed in one is still open.
Judgment calls
- If the "bug" is actually intended behavior, stop and explain before changing anything.
- If the correct fix requires a breaking change or wide refactor, present the minimal safe fix and the larger fix as options rather than choosing the invasive one unilaterally.
- If the fix touches security-sensitive code (auth, permissions, input handling), also apply the
security-auditskill's checklist to the change.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: thixpin
- Source: thixpin/claude-config
- License: MIT
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.