Verify that code changes do not introduce OAuth security vulnerabilities. Use when modifying token handling, client authentication, scope validation, redirect URI checks, secret comparison, or grant flows in Doorkeeper.
End-to-end process for handling a Rack security report (GitHub Security Advisory) from receipt to backported release. Use when asked to triage, review, accept, describe, fix, merge, or backport a security advisory, or to review a review document for one.
5.1k18d agoReviewMountCreate a free account to mount
Security audit of a codebase — web apps, APIs, services, CLI tools, libraries, daemons, and more. Use when asked to find security bugs, do a security review, audit for vulnerabilities, or pen-test the code. Focuses on exploitable issues with real impact, not theoretical concerns or industry-standard behavior.
Work the security architecture in the existing model - classify data, mark exposure and trust zones, find unguarded paths from the public internet, propose threat assessments for trust-boundary crossings, anchor high-rated systems to a control framework (NIST SP 800-53, SP 800-207, CSF), and check SBOMs against security standards. Use for security architecture reviews, threat modelling, attack-surface or "what could an attacker reach" questions, data classification, zero-trust or control-coverage questions, and SEC001-SEC003 or ALN009 findings.
Security for AI-powered software — LLM applications, agents, tool calling, RAG, MCP servers and AI-assisted coding. Covers prompt injection (direct and indirect), excessive agency, output handling, data leakage, vector-store tenant isolation, AI-provider key theft and spend abuse, supply chain for models, plugins and skills, and testing and monitoring of AI features. Use when the user says "prompt injection", "jailbreak", "secure my chatbot", "LLM security", "AI agent security", "MCP security", "tool calling safe", "RAG security", "OWASP LLM", "system prompt leak", "AI API key leaked", "unexpected OpenAI bill", "unexpected Anthropic bill", "agent permissions", "can the agent be tricked", "AI inference provider attack" or "is it safe to give the agent shell access"; when a project calls an LLM API, defines tools or agents, runs an MCP server, or loads third-party skills or plugins; and as a pass in any audit of an AI feature. Defensive security only. By Devleck.
Expert knowledge for Azure Security development including best practices, decision making, security, configuration, integrations & coding patterns, and deployment. Use when configuring Azure antimalware, container SBOM/signing, CMK encryption, AKS/VM hardening, or Key Vault keys, and other Azure Security related development tasks. Not for Azure Defender For Cloud (use azure-defender-for-cloud), Azure Information Protection (use azure-information-protection), Azure DDoS Protection (use azure-ddos-protection), Azure Web Application Firewall (use azure-web-application-firewall).
8085mo agoReviewMountCreate a free account to mount
Guideline for designing, implementing, and verifying secure APIs following OWASP API Security Top 10 (2023) best practices. Use when the user wants to: (1) review API code or design for security vulnerabilities, (2) design a secure REST, GraphQL, or gRPC API architecture, (3) implement API authentication and authorization (OAuth2, JWT, API keys, mTLS), (4) configure rate limiting, input validation, or CORS, (5) audit API endpoints for BOLA, BFLA, or mass assignment vulnerabilities, (6) create API security checklists or verification plans, (7) fix API security bugs or harden existing APIs, (8) set up API security testing (OWASP ZAP, Schemathesis, Burp Suite), or (9) handle any API security concern including SSRF prevention, resource consumption limits, business flow protection, API inventory management, and secure third-party API consumption.
Java security checklist covering OWASP Top 10, input validation, injection prevention, and secure coding. Works with Spring, Quarkus, Jakarta EE, and plain Java. Use when reviewing code security, before releases, or when user asks about vulnerabilities.
Plain text files in a repository that tell a coding agent how the project works: commands to run, conventions to follow and things to avoid. CLAUDE.md, AGENTS.md, cursor rules and skills are the common kinds.
CLAUDE.md or AGENTS.md?
CLAUDE.md is read by Claude Code. AGENTS.md is an open format that Codex, Cursor and other agents read. Many projects keep one and point the other at it.
What is a skill?
A folder with a SKILL.md that describes one capability, such as filling PDFs or reviewing code. The agent loads it only when the task calls for it.
Can I search my own team's files too?
Your agents already can, over MCP, limited to the files you're allowed to read. Searching them from this page is coming.