GOVERN Developer
Governance as code. Ship AI features with confidence.
Integration options
GOVERN provides four paths to embed governance in your development workflow:
| Option | Best for | Setup time |
|---|---|---|
| TypeScript SDK | Node.js and browser apps | 5 minutes |
| Python SDK | ML pipelines, notebooks, FastAPI | 5 minutes |
| Go SDK | High-performance services | 5 minutes |
| CLI | CI/CD pipelines, scripting | 2 minutes |
TypeScript in 30 seconds
npm install @archetypal-ai/governimport { GovernClient } from '@archetypal-ai/govern';
const govern = new GovernClient({ apiKey: process.env.GOVERN_API_KEY, orgId: process.env.GOVERN_ORG_ID,});
// Wrap any AI callconst result = await govern.assess({ model: 'claude-sonnet-4-20250514', prompt: messages, response: completion.content,});
if (result.action === 'block') { throw new GovernPolicyViolation(result.violations);}Python in 30 seconds
pip install archetypal-governfrom archetypal_govern import GovernClient
govern = GovernClient( api_key=os.environ["GOVERN_API_KEY"], org_id=os.environ["GOVERN_ORG_ID"],)
result = govern.assess( model="claude-sonnet-4-20250514", prompt=messages, response=completion.content,)CI/CD in 2 minutes
- name: GOVERN assessment uses: archetypal-ai/govern-action@v1 with: api-key: ${{ secrets.GOVERN_API_KEY }} model: claude-sonnet-4-20250514 test-prompts: tests/govern/prompts.jsonWhat you get
- Inline assessment — assess any inference before acting on it
- Batch assessment — assess test datasets in CI
- Policy enforcement — block, flag, or log based on org policy
- Webhook events — receive real-time alerts in Slack, PagerDuty, or custom endpoints
- Audit trail — every assessment logged with full metadata