Find bugs, security vulnerabilities, and code quality issues in local branch changes. Use when asked to review changes, find bugs, security review, or audit code on the current branch.
Evidence: Why this is real (not already fixed, no existing test, etc.)
Fix: Concrete suggestion
References: OWASP, RFCs, or other standards if applicable
If you find nothing significant, say so - don't invent issues.
Do not make changes - just report findings. I'll decide what to address.
1---2name: review-find-bugs3description: Find bugs, security vulnerabilities, and code quality issues in local branch changes. Use when asked to review changes, find bugs, security review, or audit code on the current branch.4---56# Find Bugs
78Review changes on this branch for bugs, security vulnerabilities, and code quality issues.
910## Phase 1: Complete Input Gathering
11121. Get the FULL diff: `git diff $(gh repo view --json defaultBranchRef --jq '.defaultBranchRef.name')...HEAD`
132. If output is truncated, read each changed file individually until you have seen every changed line
143. List all files modified in this branch before proceeding
1516## Phase 2: Attack Surface Mapping
1718For each changed file, identify and list:
1920* All user inputs (request params, headers, body, URL components)
21* All database queries
22* All authentication/authorization checks
23* All session/state operations
24* All external calls
25* All cryptographic operations
2627## Phase 3: Security Checklist (check EVERY item for EVERY file)
2829* [ ] **Injection**: SQL, command, template, header injection
30* [ ] **XSS**: All outputs in templates properly escaped?
31* [ ] **Authentication**: Auth checks on all protected operations?
32* [ ] **Authorization/IDOR**: Access control verified, not just auth?
33* [ ] **CSRF**: State-changing operations protected?
34* [ ] **Race conditions**: TOCTOU in any read-then-write patterns?
35* [ ] **Session**: Fixation, expiration, secure flags?
36* [ ] **Cryptography**: Secure random, proper algorithms, no secrets in logs?
37* [ ] **Information disclosure**: Error messages, logs, timing attacks?
38* [ ] **DoS**: Unbounded operations, missing rate limits, resource exhaustion?
39* [ ] **Business logic**: Edge cases, state machine violations, numeric overflow?
4041## Phase 4: Verification
4243For each potential issue:
4445* Check if it's already handled elsewhere in the changed code
46* Search for existing tests covering the scenario
47* Read surrounding context to verify the issue is real
4849## Phase 5: Pre-Conclusion Audit
5051Before finalizing, you MUST:
52531. List every file you reviewed and confirm you read it completely
542. List every checklist item and note whether you found issues or confirmed it's clean
553. List any areas you could NOT fully verify and why
564. Only then provide your final findings
5758## Output Format
5960**Prioritize**: security vulnerabilities > bugs > code quality
6162**Skip**: stylistic/formatting issues
6364For each issue:
6566* **File:Line** - Brief description
67* **Severity**: Critical/High/Medium/Low
68* **Problem**: What's wrong
69* **Evidence**: Why this is real (not already fixed, no existing test, etc.)
70* **Fix**: Concrete suggestion
71* **References**: OWASP, RFCs, or other standards if applicable
7273If you find nothing significant, say so - don't invent issues.
7475Do not make changes - just report findings. I'll decide what to address.
Run npx skillmds add majiayu000/review-find-bugs in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.
Find bugs, security vulnerabilities, and code quality issues in local branch changes. Use when asked to review changes, find bugs, security review, or audit code on the current branch. It is listed under Security on SkillMD.
This skill has not completed SkillMD's automated safety review yet. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.
This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.
Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.
majiayu000 (@majiayu000) published this skill. Their other Agent Skills are listed on their SkillMD profile.