a2a-security-review
Review agent-to-agent (A2A) / multi-agent-system trust: peer identity and authentication, message integrity, capability-negotiation trust, and delegation-chain…
Enrich and pivot on indicators of compromise — resolve context, infrastructure, and relationships, assess confidence and relevance, and decide block vs. monitor. Use to add analytic value to raw IOCs from an incident, feed, or hunt.
$ npx -y skills add jassics/awesome-claude-security --skill ioc-enrichment --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/ioc-enrichmentContext preview
The summary Claude sees to decide when to auto-load this skill.
Enrich and pivot on indicators of compromise — resolve context, infrastructure, and relationships, assess confidence and relevance, and decide block vs. monitor. Use to add analytic value to raw IOCs from an incident, feed, or hunt.
name: ioc-enrichment description: >- Enrich and pivot on indicators of compromise — resolve context, infrastructure, and relationships, assess confidence and relevance, and decide block vs. monitor. Use to add analytic value to raw IOCs from an incident, feed, or hunt.
IOCs turned into context: what each indicator is, how it relates to others and to known actors, how confident/relevant it is, and what defensive action it warrants.
1. **Characterize** each indicator (hash, IP, domain, URL, email, cert/JA3) and its provenance and first/last-seen. 2. **Enrich** with context: passive DNS, WHOIS/registration, hosting/ASN, certificate and infrastructure overlaps, sandbox/sample reports, reputation, and prior sightings in your environment. 3. **Pivot** — use overlaps (shared infrastructure, registrant, TLS certs, malware config) to discover related indicators and cluster activity. Map to actors/ campaigns where evidence supports it. 4. **Assess** — confidence and source reliability; filter benign/shared infrastructure (CDNs, sinkholes, common services) to avoid false positives. 5. **Decide** — block / monitor / ignore, with rationale and an expiry/review (atomic indicators age out).
An enriched indicator set: indicator · type · context/infrastructure · related indicators · actor/campaign link · confidence · action · expiry. Confirmed-malicious TTPs → `detection-engineering`; clusters → `threat-actor-profiling`.
Validate before blocking — shared/benign infrastructure on a blocklist causes outages. Pivot on durable overlaps (infrastructure, certs, malware config) rather than treating indicators in isolation. Atomic IOCs decay; set review/expiry so stale blocks don't accumulate.
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…