Sentinel API
The definitive engineering standard for deterministic AI compliance. Integrate technical evidence collection directly into your development lifecycle.
Architecture
Sentinel operates as a Zero-Egress infrastructure layer. Compliance logic is executed locally within your CI/CD runner or development environment. Your source code, model weights, and proprietary training data never leave your secure perimeter.
Sovereign Evidence Collection
By using deterministic article mapping, Sentinel translates repository signals (metadata, configs, oversight logs) into verifiable compliance artifacts.
API Audit Endpoint
curl -X POST https://api.gettingsentinel.com/api/audit \
-H "Authorization: Bearer YOUR_KEY" \
-H "Content-Type: application/json" \
-d @manifest.jsonCI/CD Integration
Run Sentinel in your GitHub Actions or GitLab pipelines to block non-compliant PRs before they hit production.
# CI/CD integration (exit 0 = COMPLIANT, exit 1 = NON_COMPLIANT)
npx @radu_api/sentinel-scan ./manifest.json && kubectl apply -f deployment.yamlVerification Badge
Display your technical compliance status directly on your repository. The Sentinel Verification Badge provides an informational, deterministic signal that your project has undergone technical auditing.
[](https://gettingsentinel.com)AI Compliance Tools
Sentinel is a deterministic compliance infrastructure designed for engineering teams building High-Risk AI systems under the EU AI Act. By integrating directly into development workflows, Sentinel provides a technical evidence layer that links repo-level signals, manifests, and rule evaluations.
The system supports zero-egress scanning, allowing rule verification to occur locally or within isolated enterprise infrastructure. To understand how deterministic scanners differ from traditional governance platforms, see the Sentinel vs AI Governance Tools comparison. For specific technical queries, refer to our Compliance FAQ and technical explainer.