ClawMaven + Cisco AI Defense

Version: 1.10.37  |  Owner: Master 22 Solutions  |  ClawMaven AI Governance Platform

This integration guide (/integrations/cisco-ai-defense) sets up the scanner_pass approval gate. A governance pack can require a clean Cisco Skill Scanner or Cisco MCP Scanner report before a skill or MCP server is promoted. You run the scanner and place the JSON report in the pack's evidence/scans/ folder; ClawMaven validates that a scan ran and passed by reading the report's findings, never the scanner's exit code.

What it is

A governance pack can require a clean Cisco Skill Scanner report for each generated skill, or a Cisco MCP Scanner report for an MCP server, before promotion. Both scanners are open source (Apache-2.0) from Cisco AI Defense. The customer runs the scanner; ClawMaven verifies the report. A clean report is best-effort detection by the scanner, not a certification.

How to turn it on

  1. In the wizard's approvals step, tick "Require a Cisco Skill Scanner / MCP Scanner clean report before promotion". It is off by default.
  2. Defaults: fail on CRITICAL or HIGH findings; maximum report age 30 days. MEDIUM and the maximum age are adjustable under Advanced.
  3. Generate the pack; it gains HANDOFF.md and evidence/scans/README.md. Install with pip install cisco-ai-skill-scanner and run the command HANDOFF.md prints, for example: skill-scanner scan skills/ --format json --output evidence/scans/.json --policy balanced
  4. Place each JSON report in evidence/scans/ and run verify/verify.sh or verify/verify.ps1.

What verify.sh checks

One PASS or FAIL row per gate, with the report's SHA-256 on PASS. A gate fails when the report is missing or unparseable, when any finding's severity is in fail_on, or when the report's timestamp is older than the maximum age. The scanner's exit code, is_safe and max_severity are never consulted.

Relationship to DefenseClaw

DefenseClaw is an open-source (Apache-2.0) runtime enforcement and evidence layer from Cisco AI Defense for OpenClaw and other agent hosts. Complementary: ClawMaven decides and documents what an agent may do before deployment; DefenseClaw enforces and logs at run time. Teams running OpenClaw can use both. See ClawMaven vs DefenseClaw, open the wizard, or book the Governance Audit.

Key Facts

Supported Runtimes (14 total)

Native, local and cloud runtimes — support level in brackets

Heuristic (Config Scout) — Governance gap analysis via code scanning

Pricing

Key Platform Features

Governance Wizard

24-step guided wizard: agent goal, operator mode, runtime target, budget caps, forbidden actions, model routing, scheduling, escalation rules, skill sources, and risk thresholds. Live Autonomy Intensity Meter (0–100) with top-3 risk drivers. Four editions: Solo, Trading, Agency, Enterprise. Five governance templates: Research, Content, DevOps, Support, Trading.

Config Scout (/optimize)

Upload or paste any existing AI agent config. Scans against 40+ heuristic patterns across all 5 heuristic runtimes (Google ADK, OpenAI Agents SDK, LangGraph, CrewAI, Claude / Anthropic SDK). Returns suggested fixes and a governance pack. Closes compliance gaps in seconds. Available in Builder and above.

Runtime Hygiene Engine (/hygiene)

7-card advisory dashboard: Session file count, session registry size, large session transcripts, orphaned session references, scheduler / cron log size, stale path references and overall cleanup confidence. Generates maintenance policies and actionable repair commands.

Auditor Compliance Summary

10-section compliance summary bundled in every governance ZIP. Covers EU AI Act, NIST AI RMF, ISO/IEC 42001, SOC 2, and GDPR. EU AI Act coverage spans Articles 9, 10, 13, 14, 17 and 62. The EU AI Omnibus (Regulation (EU) 2026/1744, in force 27 July 2026) moved stand-alone high-risk (Annex III) obligations to 2 December 2027 and product-embedded (Annex I) to 2 August 2028. Article 50 transparency obligations have applied since 2 August 2026.

Skill Assurance Pipeline (/skill-health)

5-stage validation: Stage 1 Health Scoring and Drift Detection, Stage 2 Incident Classification, Stage 3 Candidate Improvement Drafts, Stage 4 Candidate Validation, Stage 5 Promotion Gates. Stages 1–2 free; stages 3–5 require Builder or above.

Run Monitoring Dashboard (/monitor)

Your agents send their own run reports to POST /api/runs/ingest with an API key; ClawMaven does not connect to or watch your runtime. The /monitor dashboard lists the reports it receives by agent (status, last-seen time, error counts), re-fetches them every 30 seconds while Live mode is on, applies alert rules for cost spikes, error rates and budget breaches, and exports an incident summary for compliance review. Team and Enterprise.

Custom Skills Generator (/skills-generator)

Runtime-agnostic 6-step wizard for building custom AI agent skills with defined inputs, outputs, permissions, and safety rails. ZIP README includes Runtime Quick-Start install guides for OpenClaw, CrewAI, LangGraph, Agent Zero, and NullClaw. Builder/Team tiers.

OpenClaw Repair CLI (/cli)

Standalone Node.js tool (clawmaven binary) for diagnosing and repairing OpenClaw installations. Commands: diagnose, repair, install --patch-anthropic, guardian (live watchdog). Fixes 1006 gateway errors, corrupted config.json, OAuth silent failures, budget bleed.

All Public Pages

Comparison pages (11)

Full AI-readable documentation: ClawMaven full AI documentation (llms-full.txt)

Concise AI summary: ClawMaven AI summary (llms.txt)