
Claude Security Scanner
Scans your code or changes for security issues and drafts patch files you review before applying.
- Repository
- anthropics/claude-plugins-official
- Stars
- 45,770
- Licence
- Apache-2.0
- Files
- 7 files
What it does
Once installed, the agent can run a security review of a whole repository, a scoped part of it, or only the changes on a branch, pull request, or single commit. Each reported finding is checked by a separate verifier pass before it appears in a report. The agent can also turn findings into targeted patch files on disk, which you review and apply yourself; nothing is committed, pushed, or opened as a pull request.
Great for
- Developers who want a second, AI-driven security review of their own repository before merging
- Teams reviewing a pull request or branch diff for security problems
- Engineers who want suggested fixes delivered as patch files rather than automatic edits
Skip it if
- Scanning untrusted or unknown code directly, because the scan runs in your session under your permissions with no isolation layer; the instructions recommend a sandbox runtime for that case
- Replacing a dedicated, audited security tool or penetration test for compliance-critical systems, since findings come from an AI review
Try asking
- “Scan this repo for security issues”
- “Scan my branch's changes for vulnerabilities before I open a pull request”
- “Scan commit abc1234 with Claude Security”
- “Suggest patches for the Claude Security findings from the last scan”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 7 original files, the Apache-2.0 licence and an attribution note.
claude-security.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip claude-security.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip claude-security.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Open the session in the repository and enable auto mode (Shift+Tab or claude --permission-mode auto) so the scan runs with fewer interruptions.
- Scope a codebase scan to a directory when the full repository is large, and use scan changes for pull-request-sized reviews.
- Review every patch file yourself before applying it, and treat the verifier panel as an aid rather than a guarantee.
What's inside
- jobs/scan-changes.md
- jobs/scan-codebase.md
- jobs/suggest-patches.md
- role.md
- SKILL.md
- specs/patch-spec.md
- specs/report-spec.md
Redistributed unmodified from anthropics/claude-plugins-official under the Apache-2.0 licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.