a2a-security-review
Review agent-to-agent (A2A) / multi-agent-system trust: peer identity and authentication, message integrity, capability-negotiation trust, and delegation-chain…
Drive a security incident through the response lifecycle (NIST SP 800-61 / SANS PICERL): triage and scope, contain, eradicate, recover, and capture lessons learned. Use to coordinate or work an active incident. Authorized responders only.
$ npx -y skills add jassics/awesome-claude-security --skill incident-response --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/incident-responseContext preview
The summary Claude sees to decide when to auto-load this skill.
Drive a security incident through the response lifecycle (NIST SP 800-61 / SANS PICERL): triage and scope, contain, eradicate, recover, and capture lessons learned. Use to coordinate or work an active incident. Authorized responders only.
name: incident-response description: >- Drive a security incident through the response lifecycle (NIST SP 800-61 / SANS PICERL): triage and scope, contain, eradicate, recover, and capture lessons learned. Use to coordinate or work an active incident. Authorized responders only.
A controlled response that limits damage, removes the adversary, restores operations, and produces an evidence-backed record — without destroying evidence or tipping off the attacker prematurely.
1. **Preparation** — confirm authority, roles, comms plan, and tooling (mostly pre-incident; verify they're in place). 2. **Identification / triage** — validate the incident is real; determine type, scope, affected assets, and severity. Preserve volatile evidence first (`forensic-triage`). 3. **Containment** — short-term (isolate hosts, block C2, disable accounts) then long-term, balancing speed against evidence preservation and attacker awareness. 4. **Eradication** — remove the foothold: malware, persistence, created accounts, and the root cause/initial access vector. 5. **Recovery** — restore from known-good, validate integrity, monitor for return, and lift containment in a controlled way. 6. **Lessons learned** — post-incident review: timeline, root cause, what worked, and improvements (detections, controls, process).
1. Establish scope and severity; declare and track the incident. 2. Work the phases; at each step record actions, timestamps, and evidence handled. 3. Extract IOCs (`ioc-development`) and feed containment/detection in parallel. 4. Drive to root cause and confirm full eradication before recovery.
An incident record: classification · scope · timeline · actions · IOCs · root cause · recovery status · lessons. Use `security-reporting` for the incident report; recommend detections via `detection-engineering`.
Contain without destroying evidence or alerting the adversary prematurely — sequence matters. "Eradicated" means the **root cause and all persistence** are gone, not just the malware you first saw. Capture lessons into durable detections and control changes, or the next incident repeats.
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…