a2a-security-review
Review agent-to-agent (A2A) / multi-agent-system trust: peer identity and authentication, message integrity, capability-negotiation trust, and delegation-chain…
Look up the right OWASP ASVS (Application Security Verification Standard) v5.0 chapter and verification level (L1/L2/L3) for a control, requirement, or finding. Use when a design review, secure-code review, or pentest finding needs a consistent ASVS citation, or when scoping how
$ npx -y skills add jassics/awesome-claude-security --skill asvs-reference --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/asvs-referenceContext preview
The summary Claude sees to decide when to auto-load this skill.
Look up the right OWASP ASVS (Application Security Verification Standard) v5.0 chapter and verification level (L1/L2/L3) for a control, requirement, or finding. Use when a design review, secure-code review, or pentest finding needs a consistent ASVS citation, or when scoping how
name: asvs-reference description: >- Look up the right OWASP ASVS (Application Security Verification Standard) v5.0 chapter and verification level (L1/L2/L3) for a control, requirement, or finding. Use when a design review, secure-code review, or pentest finding needs a consistent ASVS citation, or when scoping how deep a verification effort should go.
The correct ASVS chapter/requirement and verification level for a control or finding, cited consistently — so design reviews, secure-code reviews, and audits anchor to the same standard.
1. **Pick the verification level** by asset risk tier:
via black-box/automated checks.
most business apps).
infra, regulatory mandate) — the full control set, deeper verification. 2. **Map the finding/control to the right ASVS chapter** — see `reference.md` for the chapter list. Cite the current ASVS chapter/requirement ID; don't hardcode a numbering you're not certain of — confirm against the live standard if precision matters (chapter numbering has shifted between ASVS editions). 3. **State the edition** — target **ASVS v5.0**; older reviews may cite v4.0.3, note the version if mixing. 4. **Hand off appropriately** — ASVS gives the control catalog; `security-architect: security-design-review` and `security-architect:secure-architecture-maturity` use it for design-time verification depth, `web-app-security`/`api-security` use it during testing.
Chapter/requirement + verification level (L1/L2/L3) + one-line rationale for the level chosen. For control-maturity framing beyond a single finding, see `security-architect:secure-architecture-maturity` (SAMM).
ASVS is a **verification checklist**, not a maturity model — for "how mature is our security program" use OWASP SAMM instead (see `security-architect: secure-architecture-maturity`). Don't over-cite L3 requirements against an L1 asset; match level to actual risk. See `reference.md` for the chapter table and level criteria.
A Claude Code plugin marketplace for the full cybersecurity & GenAI-security lifecycle — from recon and threat modeling to detection engineering, GRC, and CISO-level strategy. A pentester knows which OWASP test bends a broken-access-control endpoint.
Repo: jassics/awesome-claude-security
Review agent-to-agent (A2A) / multi-agent-system trust: peer identity and authentication, message integrity, capability-negotiation trust, and delegation-chain…
Test the agent execution harness/runtime itself — LangChain/LangGraph, AutoGen, CrewAI, custom ReAct-style loops, or computer-use/browser-use agents — for…
Assess an autonomous / tool-using AI agent for security end-to-end: tool privileges, autonomy and approval boundaries, excessive agency, memory/state…
Test what an AI agent will actually do without human confirmation, including under injected-goal / prompt-injection scenarios, to validate its autonomy and…
Review the security of MCP (Model Context Protocol) servers/clients an agent uses: server trust tier, tool/resource description and result poisoning,…
Inventory the tools/functions an AI agent can call and audit their privileges, side effects, and approval requirements to find excessive-agency and…