a2a-security-review
Review agent-to-agent (A2A) / multi-agent-system trust: peer identity and authentication, message integrity, capability-negotiation trust, and delegation-chain…
Scan a cloud environment (AWS/Azure/GCP) for high-impact misconfigurations and exposures — public storage, open ingress, unencrypted data, exposed secrets/ metadata, missing logging — and prioritize quick wins. Use for a fast exposure sweep on an authorized environment.
$ npx -y skills add jassics/awesome-claude-security --skill cloud-misconfig-scan --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/cloud-misconfig-scanContext preview
The summary Claude sees to decide when to auto-load this skill.
Scan a cloud environment (AWS/Azure/GCP) for high-impact misconfigurations and exposures — public storage, open ingress, unencrypted data, exposed secrets/ metadata, missing logging — and prioritize quick wins. Use for a fast exposure sweep on an authorized environment.
name: cloud-misconfig-scan description: >- Scan a cloud environment (AWS/Azure/GCP) for high-impact misconfigurations and exposures — public storage, open ingress, unencrypted data, exposed secrets/ metadata, missing logging — and prioritize quick wins. Use for a fast exposure sweep on an authorized environment.
A prioritized list of concrete misconfigurations, focused on the high-impact, commonly-exploited exposures that deliver the fastest risk reduction.
1. **Public data** — world-readable/writable object storage, public snapshots/ images, public databases/caches. 2. **Open network** — `0.0.0.0/0` to SSH/RDP/DB/admin ports; overly broad security groups/NSGs/firewall rules. 3. **Encryption gaps** — unencrypted storage/volumes/databases; TLS not enforced; default/unmanaged keys for sensitive data. 4. **Identity exposure** — public/cross-account grants, wildcard policies, no MFA on privileged users (cross-ref `cloud-iam-review`). 5. **Secrets & metadata** — secrets in user-data/env/code; metadata service unprotected (AWS IMDSv1 allowed); credentials reachable from compute. 6. **Logging off** — audit logging or threat detection disabled in any region/account.
1. Confirm scope and read access. Use provider config data / a CSPM tool if available. 2. Check each category; for each hit, capture the resource, exposure, and blast radius. Prioritize internet-reachable and data-exposing findings. 3. Record: category · resource · exposure · severity · fix.
A prioritized misconfig table: category · resource · exposure · severity · remediation (quick-win flag). Confirmed issues → `security-reporting:finding`.
This is the fast exposure sweep; `cloud-posture-review` is the comprehensive, CIS-mapped assessment — run this first for quick wins, then the full review. Public data + open admin ports are the highest-frequency real-world cloud incidents.
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…