How do I install the Code Review skill?

Run npx skillmds@latest add specterops/code-review 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.

Is Code Review safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Code Review?

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.

Is Code Review free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Code Review?

specterops (@specterops) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Code Review by specterops — Agent Skill

Name: Code Review
Author: specterops

File contents

Security-Focused Code Review

Review current git changes with a focus on security issues and code quality.

Parse the user's input to determine scope:

$code-review → review all changes (staged + unstaged)
$code-review staged → only staged changes
$code-review unstaged → only unstaged changes

Steps

Determine scope from user input (default: all)
- staged: git diff --staged
- unstaged: git diff
- all: both git diff and git diff --staged
Run git status for the full picture
Read the full content of each changed file for context (not just the diff)
Analyze for:

Security Issues (Critical)

Injection: SQL injection, command injection, XSS, SSTI, LDAP injection
Hardcoded secrets: API keys, passwords, tokens, connection strings, private keys
Path traversal: unsanitized file path inputs, ../ patterns
Insecure deserialization: pickle, yaml.load without SafeLoader, json with custom decoders
Weak crypto: MD5/SHA1 for security, ECB mode, hardcoded IVs, insufficient key lengths
SSRF: unvalidated URLs in HTTP requests, DNS rebinding potential
Race conditions: TOCTOU, shared state without locks, async hazards
Auth issues: missing auth checks, privilege escalation, IDOR patterns
CORS misconfiguration: overly permissive Access-Control-Allow-Origin, credentials with wildcard
JWT issues: algorithm confusion (none/HS256 vs RS256), missing expiration, weak secrets

Logic & Correctness

Off-by-one errors, boundary conditions
Unhandled error cases that could crash or leak info
Resource leaks (unclosed files, connections, sockets)
Incorrect type handling

Code Quality

Overly complex logic that could hide bugs
Missing input validation at trust boundaries
Inconsistent error handling patterns

Present findings by severity:
- CRITICAL: Exploitable security vulnerabilities
- HIGH: Security weaknesses, data exposure risks
- MEDIUM: Logic bugs, error handling gaps
- LOW: Style issues, minor improvements
Each finding includes: file:line, issue, why it matters, suggested fix
Summary: total by severity, verdict (safe to merge / needs fixes / needs rework)

---name: code-reviewdescription: Security-focused code review of current git changes. Use when reviewing code diffs, checking for vulnerabilities, or before merging branches.---# Security-Focused Code ReviewReview current git changes with a focus on security issues and code quality.Parse the user's input to determine scope:- `$code-review` → review all changes (staged + unstaged)- `$code-review staged` → only staged changes- `$code-review unstaged` → only unstaged changes## Steps1. Determine scope from user input (default: all) - staged: `git diff --staged` - unstaged: `git diff` - all: both `git diff` and `git diff --staged`2. Run `git status` for the full picture3. Read the full content of each changed file for context (not just the diff)4. Analyze for:### Security Issues (Critical)- **Injection**: SQL injection, command injection, XSS, SSTI, LDAP injection- **Hardcoded secrets**: API keys, passwords, tokens, connection strings, private keys- **Path traversal**: unsanitized file path inputs, `../` patterns- **Insecure deserialization**: pickle, yaml.load without SafeLoader, json with custom decoders- **Weak crypto**: MD5/SHA1 for security, ECB mode, hardcoded IVs, insufficient key lengths- **SSRF**: unvalidated URLs in HTTP requests, DNS rebinding potential- **Race conditions**: TOCTOU, shared state without locks, async hazards- **Auth issues**: missing auth checks, privilege escalation, IDOR patterns- **CORS misconfiguration**: overly permissive Access-Control-Allow-Origin, credentials with wildcard- **JWT issues**: algorithm confusion (none/HS256 vs RS256), missing expiration, weak secrets### Logic & Correctness- Off-by-one errors, boundary conditions- Unhandled error cases that could crash or leak info- Resource leaks (unclosed files, connections, sockets)- Incorrect type handling### Code Quality- Overly complex logic that could hide bugs- Missing input validation at trust boundaries- Inconsistent error handling patterns5. Present findings by severity: - CRITICAL: Exploitable security vulnerabilities - HIGH: Security weaknesses, data exposure risks - MEDIUM: Logic bugs, error handling gaps - LOW: Style issues, minor improvements6. Each finding includes: file:line, issue, why it matters, suggested fix7. Summary: total by severity, verdict (safe to merge / needs fixes / needs rework)

Frequently asked questions

Code Review