Adaptive Skill Assurance — 5-Stage Skill Governance Health Pipeline

Version: 1.5.12  |  Owner: Master 22 Solutions  |  ClawMaven AI Governance Platform

The Adaptive Skill Assurance pipeline (/skill-health) is a five-stage, client-side pipeline for evaluating AI agent skill governance health and generating improvement recommendations. It runs entirely in your browser and processes skill definitions against governance criteria without uploading data to any server.

The 5 Assurance Stages

1. Stage 1: Health Scoring — Evaluate each skill against governance criteria (permission declarations, safety rails, input/output schema completeness). Free on Starter.
2. Stage 2: Drift Detection — Compare current skill definitions against their baseline to detect undocumented changes. Free on Starter.
3. Stage 3: Active Incident Classification — Classify governance incidents associated with skills in production. Requires Builder or above.
4. Stage 4: Candidate Improvement Drafts — Generate improvement recommendations for skills that fail health checks. Requires Builder or above.
5. Stage 5: Promotion Gates — Enforce promotion criteria before skills advance to production. Requires Builder or above.

Key Facts

• Owner: Master 22 Solutions (contact@master22solutions.com)
• Website: https://clawmaven.com
• Current version: 1.5.12
• Supported runtimes: 11 (5 Native + 6 Heuristic)
• Privacy: wizard config stays in the browser — nothing transmitted to servers
• Free to start — no credit card required, no account needed
• Compliance: EU AI Act (August 2026), NIST RMF, ISO 42001

Supported Runtimes (11 total)

Native — Full governance pack generation

• OpenClaw v1.x — Open-source agent framework; full native governance config + Repair CLI integration
• Perplexity API — Browser-control integration; governance artifacts mapped to Perplexity's permission model
• LM Studio — Local / air-gapped models; routes tasks to local LLM server with three routing presets
• Agent Zero — Modular multi-agent setups; markdown-based governance artifacts (instructions, prompts, agents.json)
• NullClaw — Custom runtime scaffold; 678 KB binary, <2ms startup, ideal for edge deployments

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

• LangGraph — Detects missing interrupt gates, unchecked shell execution, absent checkpointers
• CrewAI — Flags unchecked delegation chains, code execution risks in tool definitions, raw API key exposure
• Hermes — Local inference server; tool use policy analysis, prompt injection risk scoring
• Google ADK — SDK staleness detection, shell execution patterns, tool approval gap scanning
• OpenAI Agents SDK — Tool guardrail gaps, handoff governance issues, model pinning analysis
• Claude / Anthropic — Computer-use risk detection, system prompt quality scoring, tool use governance gap analysis

Pricing

• Developer (CLI) — Free. OpenClaw Repair CLI (clawmaven npm package). No account required.
• Starter — Free forever. Full 23-step wizard, all ZIP artifacts, Hygiene tools, basic monitoring, Skill Assurance.
• Builder — $49 one-time (no subscription). Config Scout with AI-pattern detection, Custom Skills Generator, Trading guardrails, full Auditor Summary exports.
• Team — $99/month. Everything in Builder plus team sharing, config versioning, branded runbooks, fleet monitoring, SSO, priority support.
• Enterprise — From $399/month or $4,500/year. Dedicated onboarding, volume licensing, custom compliance exports, white-glove support.

Key Platform Features

Governance Wizard

23-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 45+ heuristic patterns across all 6 heuristic runtimes. 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: Gateway Health, Auth Validation, Config Integrity, Budget Posture, Dependency Audit, Log Retention, Network Exposure. Generates maintenance policies and actionable repair commands.

Auditor Compliance Summary

12-section compliance summary bundled in every governance ZIP. Covers EU AI Act (Articles 9, 10, 13, 14, 17, 62), NIST RMF, ISO 42001, SOC 2 Type II, ISO 27001 Annex A. Full enforcement deadline: August 2026.

Skill Assurance Pipeline (/skill-health)

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

SaaS Fleet Monitoring (/monitor)

Live event ingest from running agents (30-second polling, alert thresholds). Fleet view with agent status, last-seen timestamps, error counts. Alert rules for cost spikes, error rates, budget breaches. Exportable monitoring logs for compliance review.

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

• / — Home: landing page, runtime coverage, 5-step process, governance templates, FAQ, pricing
• /pricing — Pricing: 5 tiers (Developer, Starter, Builder, Team, Enterprise)
• /about — About ClawMaven and Master 22 Solutions
• /security — Security practices and Privacy by Architecture architecture
• /trust-center — Trust Center: skill registry, SHA-256 hash verifier
• /help — Help Center: FAQs, Repair CLI instructions, governance concepts
• /cli — OpenClaw Repair CLI: install, command reference, diagnose output
• /optimize — Config Scout: scan existing agent configs against 45+ heuristic patterns
• /hygiene — Runtime Hygiene Engine: 7-card advisory, maintenance policies
• /validate — Config Validator: ZIP upload, schema and hash verification
• /skills-generator — Custom Skills Generator (Builder/Team tiers)
• /compare/clawmaven-vs-credo-ai — vs Credo AI (ML model risk management)
• /compare/clawmaven-vs-witness-ai — vs Witness AI (runtime anomaly detection)
• /compare/clawmaven-vs-protect-ai — vs Protect AI (ML model security scanning)
• /compare/clawmaven-vs-lakera-guard — vs Lakera Guard (LLM prompt/output filtering)
• /compare/clawmaven-vs-guardrails-ai — vs Guardrails AI (LLM output validation)
• /compare/clawmaven-vs-langsmith — vs LangSmith (LLM observability/tracing)
• /compare/clawmaven-vs-sysdig — vs Sysdig (cloud workload security)
• /compare/clawmaven-vs-microsoft-defender — vs Microsoft Defender for AI (Azure AI monitoring)

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

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