Agent skills
Security Claude skills and agent skills
What security skills do
Security reviews, static analysis, threat models and setting up auth providers correctly. A skill is a folder with a SKILL.md: the agent reads its one-line description up front and loads the instructions only when a task matches, so a security skill costs almost nothing until you need it. The 18 here come from OpenAI, Google, Trail of Bits, Sentry, Firebase (Google), Better Auth, Clerk, Auth0 (Okta), Upstash, GitHub (community contributions).
- Security Best Practices Review · OpenAI
Perform language and framework specific security best-practice reviews and suggest improvements.
- Security Threat Model · OpenAI
Repository-grounded threat modeling that enumerates trust boundaries, assets, attacker capabilities, abuse paths, and mitigations, and writes a concise Markdown threat model.
- Sign In With Google (Web) · Google
Implement, configure, and secure Sign In With Google (SiwG) using Google Identity Services (GIS / `https://accounts.google.com/gsi/client`) across web architectures.
All security & auth skills
18 of 18 skills
- Security Best Practices ReviewOfficial · OpenAI
Reviews code for language- and framework-specific security issues and suggests safer defaults.
★ 28k (repo)
Install (any agent)Claude Code, Codex, Copilot, Cursor, Gemini CLI steps →npx skills add openai/skills --skill security-best-practices - Security Threat ModelOfficial · OpenAI
Writes a threat model for a repository: trust boundaries, assets, attack paths and mitigations.
★ 28k (repo)
Install (any agent)Claude Code, Codex, Copilot, Cursor, Gemini CLI steps →npx skills add openai/skills --skill security-threat-model - Sign In With Google (Web)Official · Google
Implements and secures Google sign-in buttons and One Tap with Google Identity Services.
★ 22k (repo)
Install (any agent)Claude Code, Codex, Copilot, Cursor, Gemini CLI steps →npx skills add google/skills --skill sign-in-with-google-web - Agentic Actions AuditorOfficial · Trail of Bits
Audits GitHub Actions workflows that run AI agents for prompt-injection and privilege risks.
★ 7.8k (repo)
Install (any agent)Claude Code, Codex, Copilot, Cursor, Gemini CLI steps →npx skills add trailofbits/skills --skill agentic-actions-auditor - CodeQL ScanOfficial · Trail of Bits
Builds a CodeQL database and runs data-flow and taint analysis to find vulnerabilities.
★ 7.8k (repo)
Install (any agent)Claude Code, Codex, Copilot, Cursor, Gemini CLI steps →npx skills add trailofbits/skills --skill codeql - Constant-Time AnalysisOfficial · Trail of Bits
Finds timing side channels in cryptographic code, such as branches on secret values.
★ 7.8k (repo)
Install (any agent)Claude Code, Codex, Copilot, Cursor, Gemini CLI steps →npx skills add trailofbits/skills --skill constant-time-analysis - Differential Security ReviewOfficial · Trail of Bits
Security review of a diff that weighs blast radius, call sites and test coverage of changed code.
★ 7.8k (repo)
Install (any agent)Claude Code, Codex, Copilot, Cursor, Gemini CLI steps →npx skills add trailofbits/skills --skill differential-review - Semgrep Rule CreatorOfficial · Trail of Bits
Writes custom Semgrep rules to catch specific bug or vulnerability patterns.
★ 7.8k (repo)
Install (any agent)Claude Code, Codex, Copilot, Cursor, Gemini CLI steps →npx skills add trailofbits/skills --skill semgrep-rule-creator - Sharp EdgesOfficial · Trail of Bits
Flags error-prone APIs, risky defaults and footgun designs that invite security mistakes.
★ 7.8k (repo)
Install (any agent)Claude Code, Codex, Copilot, Cursor, Gemini CLI steps →npx skills add trailofbits/skills --skill sharp-edges - Security Review (Sentry)Official · Sentry
Checks code for injection, XSS, auth and authorization flaws and other OWASP issues.
★ 1.1k (repo)
Install (any agent)Claude Code, Codex, Copilot, Cursor, Gemini CLI steps →npx skills add getsentry/skills --skill security-review - Firebase Security Rules AuditorOfficial · Firebase (Google)
Audits Firestore and Storage rules for privilege escalation, bypasses and missing checks.
★ 485 (repo)
Install (any agent)Claude Code, Codex, Copilot, Cursor, Gemini CLI steps →npx skills add firebase/agent-skills --skill firebase-security-rules-auditor - Better Auth Best PracticesOfficial · Better Auth
Configures Better Auth server and client, database adapters, sessions and plugins.
★ 225 (repo)
Install (any agent)Claude Code, Codex, Copilot, Cursor, Gemini CLI steps →npx skills add better-auth/skills --skill better-auth-best-practices - Create Auth (Better Auth)Official · Better Auth
Scaffolds sign-in for a TypeScript app with Better Auth, OAuth providers and auth pages.
★ 225 (repo)
Install (any agent)Claude Code, Codex, Copilot, Cursor, Gemini CLI steps →npx skills add better-auth/skills --skill create-auth - Clerk Next.js PatternsOfficial · Clerk
Clerk patterns for Next.js middleware, Server Actions and caching.
★ 89 (repo)
Install (any agent)Claude Code, Codex, Copilot, Cursor, Gemini CLI steps →npx skills add clerk/skills --skill clerk-nextjs-patterns - Clerk SetupOfficial · Clerk
Adds Clerk authentication to a project, or migrates an existing auth system to Clerk.
★ 89 (repo)
Install (any agent)Claude Code, Codex, Copilot, Cursor, Gemini CLI steps →npx skills add clerk/skills --skill clerk-setup - Auth0Official · Auth0 (Okta)
Adds or fixes Auth0 login, sessions, tokens, MFA and API protection in an app.
★ 54 (repo)
Install (any agent)Claude Code, Codex, Copilot, Cursor, Gemini CLI steps →npx skills add auth0/agent-skills --skill auth0 - Upstash Rate Limit (JS)Official · Upstash
Adds rate limiting to API routes and middleware with @upstash/ratelimit.
★ 34 (repo)
Install (any agent)Claude Code, Codex, Copilot, Cursor, Gemini CLI steps →npx skills add upstash/skills --skill upstash-ratelimit-js - GitHub Actions HardeningCommunity · GitHub (community contributions)
Reviews workflow files for script injection, risky triggers and over-broad permissions.
★ 40k (repo)
Install (any agent)Claude Code, Codex, Copilot, Cursor, Gemini CLI steps →npx skills add github/awesome-copilot --skill github-actions-hardening
FAQ
What are security agent skills?+
Security reviews, static analysis, threat models and setting up auth providers correctly. Each is a SKILL.md folder that Claude Code, Codex, Cursor, Copilot and other agents load when a task matches its description. This page lists 18, from OpenAI, Google, Trail of Bits, Sentry, Firebase (Google), Better Auth, Clerk, Auth0 (Okta) and others.
What are the best security agent skills?+
Start with the ones from the publisher of the tool you use: Security Best Practices Review (OpenAI), Security Threat Model (OpenAI), Sign In With Google (Web) (Google), Agentic Actions Auditor (Trail of Bits), CodeQL Scan (Trail of Bits), Constant-Time Analysis (Trail of Bits).
How do I install a security skill?+
Run npx skills add owner/repo --skill name and pick your agent, or copy the skill folder into ~/.claude/skills (Claude Code), .agents/skills (Codex) or .github/skills (Copilot). Each skill page has the exact command for Security Best Practices Review and the rest.
More skills: Documents · Frontend & design · Testing & QA · Backend & databases · Cloud & DevOps · AI & ML · Workflow & planning · Writing & notes · Media · Data & analytics · Mobile · Payments · Integrations · MCP servers