Security workflows · commands · 18 picks

Claude Code security workflows

Security commands, hooks, rules, MCP servers, and skills for auditing code, checking risky changes, and tightening agent behavior.

Curated by @heyclaude-editors Updated 2026-07-18

Security commands, hooks, rules, MCP servers, and skills for auditing code, checking risky changes, and tightening agent behavior.

Compared at a glance

The top 5 picks side by side on trust, install, platform support, and disclosed notes — full rationale for each below.

2 trust signals differ across this comparison (Package trust, Source provenance).

Next steps differ across picks — use the actions in the table below to copy install commands and source links per resource.

Field	Prompt Injection Defense Guardrails Skill Build layered defenses against prompt injection, data exfiltration, and unsafe tool execution in AI agent systems. Open dossier	/security - Vulnerability Scan Command for Claude Code Comprehensive security audit with vulnerability detection, threat analysis, and automated remediation recommendations Open dossier	Security Auditor Expert - CLAUDE.md Rules for Claude Code Configure Claude as a security expert for vulnerability assessment, penetration testing, and security best practices Open dossier	Security Scanner Hook - Hooks Automated security vulnerability scanning that integrates with development workflow to detect and prevent security issues before deployment. Open dossier	Code Review Automation Capability Pack Skill Expert code-review capability pack for deterministic PR audits, risk-ranked findings, and low-noise fix planning without SaaS lock-in. Open dossier
Next stepsDiffers	Open dossier API JSON Open LLM Open source Newsletter Claim listing	Open dossier API JSON Open LLM Open source Newsletter Claim listing	Open dossier API JSON Open LLM Open source Newsletter Claim listing	Open dossier API JSON Open LLM Open source Newsletter Claim listing	Open dossier API JSON Open LLM Open source Newsletter Claim listing
Trust
Review status	ReviewedMaintainer reviewed	ReviewedMaintainer reviewed	ReviewedMaintainer reviewed	ReviewedMaintainer reviewed	ReviewedMaintainer reviewed
Package trustDiffers	Package verified2026-04-10	Package not verified	Package not verified	Package not verified	Package verified2026-04-11
Source provenanceDiffers	No submission link	Source-backed	Source-backed	Source-backed	No submission link
Submitter	—	—	—	—	—
Install risk	Low risk	Review first	Review first	Review first	Low risk
Notes	Safety ✓ Privacy ✓	Safety ✓ Privacy ✓	Safety ✓ Privacy ✓	Safety ✓ Privacy ✓	Safety ✓ Privacy ✓
Brand	—	—	—	—	—
Category	skills	commands	rules	hooks	skills
Source	First-party	Source-backed	Source-backed	Source-backed	First-party
Author	JSONbored	JSONbored	JSONbored	JSONbored	JSONbored
Added	2026-04-10	2025-09-16	2025-09-15	2025-09-16	2026-04-11
Platforms	Claude Code Codex Windsurf Gemini Cursor CLI	Claude Code	Claude Code	Claude Code	Claude Code Codex Windsurf Gemini Cursor CLI
Harness	Claude Code Codex Windsurf Gemini Cursor CLI	Claude Code	Claude Code	Claude Code	Claude Code Codex Windsurf Gemini Cursor CLI
Source repo	—	—	—	—	—
Safety notes	✓Treat findings and generated mitigations as review inputs; validate suspected vulnerabilities and patches before changing production controls. Coordinate disclosure-sensitive security work privately and avoid running destructive probes outside authorized systems.	✓Review generated changes and commands before applying them; slash commands can ask the agent to read, write, edit, or run tools in the current project. Limit scope to the intended files and run in a trusted checkout when the command analyzes code, tests, security findings, or generated output.	✓Only assess, scan, or test systems you own or are explicitly authorized to test; unauthorized penetration testing or exploitation is illegal. Treat any active scanning, exploitation, or DAST tooling as potentially destructive; run it against staging or scoped targets, never production without written authorization. Vulnerability findings and exploit details are sensitive; handle and disclose them responsibly rather than committing live exploits or unredacted reports.	✓Runs automatically after write, edit, or multiedit activity and invokes installed security tools against the touched file. May run truffleHog, Semgrep, npm audit, Bandit, or gosec depending on the file type and tools available. Treats scanner findings as warnings and does not guarantee that code is safe or unsafe.	✓Installs from a downloaded package and may run local commands or scaffold files as part of the workflow; review the package and any generated changes before applying.
Privacy notes	✓Security prompts and reports can include source snippets, vulnerability details, internal URLs, dependency metadata, and operational context. Redact secrets, customer data, exploit details, private infrastructure names, and unreleased findings before sharing outputs publicly.	✓Prompts, source files, logs, errors, dependency metadata, and generated reports may be sent to the configured AI model during command execution. Redact secrets, customer data, private repository details, and proprietary code before sharing command output outside the workspace.	✓Security review reads source code, configuration, environment files, and logs that can contain secrets, API keys, tokens, credentials, and PII. Do not paste discovered secrets, customer data, or internal log contents into shared chats, issues, or public notes; redact before reporting. Scanned outputs and incident artifacts may carry user data subject to GDPR/CCPA; store and transmit them only through approved, access-controlled channels.	✓Reads modified file contents for security scanning and prints scanner output to local hook output. Semgrep and package audit commands may contact external rule, registry, or advisory services depending on local tool configuration. Scanner output may include file paths, code snippets, dependency names, and suspected secrets.	✓Operates on your local project files and any context you share in the session; review what you expose before sharing — nothing is sent beyond the model unless a step calls an external service.
Prerequisites	Agent flow that processes user input or external web/content data Ability to add policy checks before tool execution Test harness for adversarial prompt scenarios	— none listed	— none listed	— none listed	Repository access and PR diff Security severity policy Test/lint/typecheck commands
Install	`curl -L https://heyclau.de/downloads/skills/prompt-injection-defense-guardrails.zip -o prompt-injection-defense-guardrails.zip && unzip -o prompt-injection-defense-guardrails.zip -d ./prompt-injection-defense-guardrails`	`/security [options] <file_or_project>`	—	`mkdir -p .claude/hooks && touch .claude/hooks/security-scanner-hook.sh && chmod +x .claude/hooks/security-scanner-hook.sh`	`curl -L https://heyclau.de/downloads/skills/coderabbit-lite-pr-review-capability-pack.zip -o coderabbit-lite-pr-review-capability-pack.zip && unzip -o coderabbit-lite-pr-review-capability-pack.zip -d ./coderabbit-lite-pr-review-capability-pack`
Config	—	—	—	`{ "hooks": { "postToolUse": { "script": "./.claude/hooks/security-scanner-hook.sh", "matchers": [ "write", "edit", "multiedit" ] } } }`	—
Citations	Source repositorygithub.com 2026-07-18T19:14:44+00:00 Documentationgenai.owasp.org Package (SHA-256 pinned)/downloads/skills/prompt-injection-defense-guardrails.zip Source methodology →	Source repositorygithub.com 2026-07-18T19:14:44+00:00 Documentationcode.claude.com Source methodology →	Source repositorygithub.com 2026-07-18T19:14:44+00:00 Documentationowasp.org Source methodology →	Source repositorygithub.com 2026-07-18T19:14:44+00:00 Documentationowasp.org Source methodology →	Source repositorygithub.com 2026-07-18T19:14:44+00:00 Documentationdocs.github.com Package (SHA-256 pinned)/downloads/skills/coderabbit-lite-pr-review-capability-pack.zip Source methodology →
Claim	Unclaimed	Unclaimed	Unclaimed	Unclaimed	Unclaimed