a2a-security-review
Review agent-to-agent (A2A) / multi-agent-system trust: peer identity and authentication, message integrity, capability-negotiation trust, and delegation-chain…
Map a finding, control, or requirement across security frameworks — CWE, NIST CSF & SP 800-53, CIS Controls, ISO/IEC 27001 — so one piece of work can be expressed in whichever framework the audience uses. Use when a finding needs a CWE, or when aligning controls/gaps across
$ npx -y skills add jassics/awesome-claude-security --skill framework-mapping --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/framework-mappingContext preview
The summary Claude sees to decide when to auto-load this skill.
Map a finding, control, or requirement across security frameworks — CWE, NIST CSF & SP 800-53, CIS Controls, ISO/IEC 27001 — so one piece of work can be expressed in whichever framework the audience uses. Use when a finding needs a CWE, or when aligning controls/gaps across
name: framework-mapping description: >- Map a finding, control, or requirement across security frameworks — CWE, NIST CSF & SP 800-53, CIS Controls, ISO/IEC 27001 — so one piece of work can be expressed in whichever framework the audience uses. Use when a finding needs a CWE, or when aligning controls/gaps across compliance frameworks.
A finding or control expressed consistently in the frameworks the audience cares about — a CWE for the bug, and the corresponding control IDs for GRC/audit — without forcing bad mappings.
1. **Classify the root cause**, not the symptom — a weakness type maps to **CWE** (e.g., SQL injection → CWE-89; missing authz → CWE-862). That's the engineering anchor. 2. **Translate to the target framework(s)**:
for program-level framing.
See `reference.md` for the families and common crosswalks. 3. **Map deliberately** — use authoritative crosswalks where they exist (e.g., 800-53 ↔ CSF, CIS ↔ multiple). Note mapping confidence; flag "no clean equivalent" rather than inventing one. 4. **For a *defensive control* citation** (not just a weakness/compliance mapping), map to **MITRE DEF3ND** — cite the technique + the countermeasure it defends against, cross-referencing `security-knowledge:attack-lookup` for the ATT&CK side. 5. **Cite versions** — frameworks revise (CSF 1.1→2.0, ISO 27001:2013→2022); state the edition.
A mapping row per item: root cause · CWE · NIST CSF · 800-53 family/control · CIS Control · ISO 27001 Annex A · DEF3ND countermeasure (if a defensive control is being specified) · confidence. For gap assessments and registers route to `grc`; for exec framing to `ciso-toolkit`/`security-reporting`.
Map root cause, not symptom — the CWE for "what's actually wrong" drives every other mapping. Crosswalks are approximate; over-precise control mappings mislead auditors. Always cite the framework edition. DEF3ND complements ATT&CK: ATT&CK says what an attacker does, DEF3ND says what countermeasure defends against it — use both when a detection/response control needs justifying. See `reference.md` for families and crosswalk pointers.
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…