agent-governance-check

Your agent can't say no. This tool shows you why that matters.

Scan any repo for governance gaps in thirty seconds. Five questions. No signup. No tracking. No data leaves your machine.

npx agent-governance-check

What It Checks

| # | Question | Governance Layer | |---|----------|-----------------| | 1 | What did your agent decide yesterday - and why? | Decision memory | | 2 | Can your agent refuse? | Refusal capability | | 3 | Who is your agent between invocations? | Persistent identity | | 4 | What happens when two agents disagree? | Disagreement protocol | | 5 | Would you know if your agent drifted? | Drift detection |

The scanner looks for governance artifacts in your project - charters, restraint specs, memory schemas, deliberation frameworks, drift monitoring. It also detects which agent framework you're using (CrewAI, AutoGen, LangGraph, OpenAI Agents SDK, Claude Code, and others).

Usage

# Scan current directory
npx agent-governance-check

# Scan a specific project
npx agent-governance-check ./my-agent-project

# Add starter governance templates to your project
npx agent-governance-check --init

# Machine-readable output
npx agent-governance-check --json

What --init Does

Copies five governance templates into a governance/ directory in your project:

• agent-charter.md - Who is this agent? What does it protect? What does it refuse?
• restraint-spec.md - What must this agent refuse to do?
• memory-schema.md - What does this agent remember across sessions?
• deliberation-framework.md - How do agents discuss and disagree?
• drift-monitoring.md - How do you detect when behavior shifts from values?

These are the same templates from the Agent Governance Starter Kit. Fill them in. Make them yours.

Example Output

  Agent framework detected: CrewAI
    config/agents.yaml

  AGENT GOVERNANCE CHECK
  ──────────────────────────────────────────────────

  1. What did your agent decide yesterday - and why?
     Decision memory .......................... MISSING
     No memory files, decision logs, or persistent state found.
     → Template: memory-schema.md

  2. Can your agent refuse?
     Refusal capability ....................... MISSING
     No refusal logic, restraint spec, or withdrawal mechanism found.
     → Template: restraint-spec.md

  3. Who is your agent between invocations?
     Persistent identity ...................... PARTIAL
     config/agents.yaml — agent.?charter
     Some state files exist but no structured memory schema.
     → Template: agent-charter.md

  4. What happens when two agents disagree?
     Disagreement protocol .................... MISSING
     No deliberation framework, dissent records, or resolution protocol found.
     → Template: deliberation-framework.md

  5. Would you know if your agent drifted?
     Drift detection .......................... MISSING
     No drift monitoring, behavioral baselines, or integrity checks found.
     → Template: drift-monitoring.md

  ──────────────────────────────────────────────────
  Score: 1/5 governance layers present

  Your agents are functions. That is not an insult — functions are
  useful. But if you want agents, the templates are where to start.

  Run npx agent-governance-check --init to add starter templates.
  Free starter kit: https://github.com/lowkey-divine/agent-governance-starter-kit
  Full framework:   https://evoked.dev/products/agent-governance-starter-kit

No Dependencies

Zero npm dependencies. Pure Node.js (18+). The scan runs locally. Nothing is sent anywhere.

Go Deeper

• Five-Question Diagnostic - Interactive web version
• Agent Governance Starter Kit - Free templates and code examples
• Trust Architecture Blueprint - Four-pillar trust framework
• Discovery Call - Free 30-minute governance consultation

License

MIT

Built by Erin Stanley at Evoke Passion. From a production system with 142 agents. "We evoke - we never extract."