a2a-security-review
Review agent-to-agent (A2A) / multi-agent-system trust: peer identity and authentication, message integrity, capability-negotiation trust, and delegation-chain…
Produce a security architecture, network, or trust-boundary diagram of a system, highlighting components, zones, controls, and exposure. Use when documenting a design review, network segmentation, or cloud architecture from a security perspective.
$ npx -y skills add jassics/awesome-claude-security --skill architecture-diagram --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/architecture-diagramContext preview
The summary Claude sees to decide when to auto-load this skill.
Produce a security architecture, network, or trust-boundary diagram of a system, highlighting components, zones, controls, and exposure. Use when documenting a design review, network segmentation, or cloud architecture from a security perspective.
name: architecture-diagram description: >- Produce a security architecture, network, or trust-boundary diagram of a system, highlighting components, zones, controls, and exposure. Use when documenting a design review, network segmentation, or cloud architecture from a security perspective.
A diagram that makes the security-relevant structure legible: components, the zones/segments they sit in, the controls between them, and where the system is exposed.
1. Gather the components (services, gateways, data stores, identities) and how they connect. Ask for the design if not provided. 2. Group into **zones** (internet, edge/DMZ, app tier, data tier, management, each cloud account/VPC/namespace). Draw zone boundaries explicitly. 3. Annotate **controls** on the links: TLS, authn/authz, WAF, firewall rules, network policy, encryption-at-rest on stores. 4. Mark **exposure**: anything internet-reachable, public buckets, admin planes. 5. Render: prefer the Excalidraw MCP; otherwise emit Mermaid `flowchart` with `subgraph` per zone, plus DOT for tooling that prefers Graphviz.
save/export it; tell the user the file location.
`.excalidraw` JSON block the user can import at excalidraw.com, (c) for graphs, a ```dot block. Keep node labels short.
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…