a2a-security-review
Review agent-to-agent (A2A) / multi-agent-system trust: peer identity and authentication, message integrity,…
Perform forensic triage on a host or artifacts — collect and analyze disk, memory, and log evidence with proper handling, then build an incident timeline. Use to investigate a compromised system or scope an incident. Preserve evidence integrity.
Claude Code
$ npx -y skills add jassics/awesome-claude-security --skill forensic-triage --agent claude-code
Codex
$ npx skills add jassics/awesome-claude-security --skill forensic-triage -a codex -y
Cursor
$ npx skills add jassics/awesome-claude-security --skill forensic-triage -a cursor -y
Gemini CLI
$ npx skills add jassics/awesome-claude-security --skill forensic-triage -a gemini-cli -y
GitHub Copilot
$ npx skills add jassics/awesome-claude-security --skill forensic-triage -a github-copilot -y
OpenCode
$ npx skills add jassics/awesome-claude-security --skill forensic-triage -a opencode -y
Windsurf
$ npx skills add jassics/awesome-claude-security --skill forensic-triage -a windsurf -y
name: forensic-triage description: >- Perform forensic triage on a host or artifacts — collect and analyze disk, memory, and log evidence with proper handling, then build an incident timeline. Use to investigate a compromised system or scope an incident. Preserve evidence integrity.
A defensible analysis of the available evidence that answers what happened, when, and how far it spread — with a timeline and chain of custody intact.
credentials, command lines (capture early — most volatile).
systemd/cron (Linux), browser/registry artifacts, deleted-file recovery.
firewall/netflow; central SIEM.
1. **Preserve first** — respect order of volatility (see `incident-response` reference); work on copies; hash and log every item (chain of custody). 2. **Triage** — quick wins: suspicious processes, persistence, recent logons, new accounts, anomalous network. Identify the foothold. 3. **Build a timeline** — normalize timestamps (UTC), correlate across sources into a single chronology of attacker actions. 4. **Scope** — pivot on IOCs/accounts/hosts to find lateral movement and the full set of affected systems; estimate dwell time. 5. **Map to ATT&CK** — tag observed techniques for clarity and detection follow-up.
A triage report: evidence collected (with hashes) · key artifacts · timeline · scope/ dwell time · ATT&CK techniques · IOCs (hand to `ioc-development`). Use `security-reporting`; visualize the timeline/attack path with `security-diagramming`.
Preserve before you analyze — and capture memory early, it's gone on reboot. A correct, source-correlated **timeline** in UTC is the backbone of the investigation. Maintain chain of custody if the findings might be used in legal/HR proceedings.
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.
Review agent-to-agent (A2A) / multi-agent-system trust: peer identity and authentication, message integrity,…
Test the agent execution harness/runtime itself — LangChain/LangGraph, AutoGen, CrewAI, custom ReAct-style loops, or…
Assess an autonomous / tool-using AI agent for security end-to-end: tool privileges, autonomy and approval boundaries, excessive…
Test what an AI agent will actually do without human confirmation, including under injected-goal / prompt-injection scenarios, to…
Review the security of MCP (Model Context Protocol) servers/clients an agent uses: server trust tier, tool/resource description…
Inventory the tools/functions an AI agent can call and audit their privileges, side effects, and approval requirements to find…
Use for authorized security assessment of REST, GraphQL, WebSocket, or SOAP APIs, including discovery, authentication,…
Extract and decrypt Windows DPAPI-protected secrets (Credential Manager, browser logins/cookies, Wi-Fi credentials, KeePass keys)…
Clean and finalize authorized natural-language text intended for readers by auditing suspicious invisible Unicode and rewriting…