security-auditor · gupsammy — Claude Code subagent: vulnerability audit
security-auditor finds exploitable vulnerabilities in auth, untrusted-input handling, secrets, and dependencies. Triggered after changes to auth, API endpoints, or file operations.
no license ★ 276 updated 2026-10-29
/plugin marketplace add gupsammy/Claudest
/plugin install claude-coding@ClaudestWhat security-auditor does
The subagent operates in two modes: quick advice during development and thorough audit of finished code. It checks for injection flaws (SQL, command, path), missing authorization, credential leaks, and outdated libraries with known CVEs. It traces how user input reaches dangerous operations by reading source files — it does not run exploits or execute project code.
It uses Read, Grep, Glob for static analysis and Bash only for git diff, find, and read‑only scanners like npm audit. If dependencies require FFmpeg, Python, or external libraries for auditing, the subagent will not work without them. WebSearch is used to verify current CVE data when a suspect package version is found.
In audit mode, it starts by mapping the attack surface: entry points, trust boundaries, secret storage. Then it traces the path from untrusted source to dangerous sink and provides path:line references for each finding.
What it helps with
- detect SQL injection and command injection in API handlers
- verify authorization before object access (IDOR)
- find hardcoded keys and tokens in code
- identify outdated dependencies with known CVEs
- trace user input path to dangerous operations
Example prompts for the agent
- Check this endpoint for SQL injection
- Run a security audit after changes to
auth.js - Is the
fileUploadhandling inroutes/api.tssafe?
How to install security-auditor
1Plugin marketplace
/plugin marketplace add gupsammy/Claudest/plugin install claude-coding@Claudest2Manually: put the files in place
.claude/agents/security-auditor.mdFor all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention
Manually: put the files in place
.codex/agents/security-auditor.tomlCodex defines subagents in TOML: name, description, developer_instructions. Move the Markdown text into developer_instructions by hand. For all projects use ~/.codex/agents/
Manually: put the files in place
.cursor/agents/security-auditor.mdCursor also reads .claude/agents/
Step by step, with commands for each platform: Claude Code subagents: how to create one, call it and keep context clean
Questions about security-auditor
How do I install security-auditor in Claude Code?
Run /plugin marketplace add gupsammy/Claudest, then /plugin install claude-coding@Claudest. The subagent will appear in .claude/agents/security-auditor.md.
What does security-auditor do?
It finds exploitable vulnerabilities: injection, missing authorization, secret leaks, unsafe dependencies. It traces the path from untrusted input to dangerous operation by reading code statically.
How is it different from code-auditor and performance-auditor?
security-auditor looks only for security issues (injection, auth, secrets). Code quality is handled by code-auditor, performance by performance-auditor, architecture by architecture-auditor.
Similar subagents
All subagents for Security →security-auditor by other authors comes first.