VeriSigil AI: SSL‑Like Cryptographic Identity for Autonomous Agents
Call the issue‑test endpoint to obtain a test passport and verify your agent with the verify endpoint.
Send a GET request to https://verisigil-api-production.up.railway.app/issue-test to receive a test passport.
Summary
VeriSigil AI provides a trust layer for autonomous AI agents, issuing a cryptographic identity passport signed with Ed25519 and including fields such as agent_id, did, signature, status, trust_score, and EU AI Act compliance. The passport is publicly verifiable and immutable, enabling audit trails, while a dynamic trust network records independent verifications, each signed and stored immutably. API endpoints allow issuing a test passport, verifying an agent, viewing the trust graph, and retrieving the DID document. The open‑source SDK is hosted on GitHub, and future features include behavioral fingerprinting, a zero‑knowledge compliance engine, an MCP security scanner, and a federated trust network. EU AI Act enforcement starts in August 2026, making this solution urgent for high‑risk AI deployments. Adopting VeriSigil ensures agents have verifiable identity, auditability, and compliance from day one.
Key changes
- Generates a DID passport signed with Ed25519 and includes agent_id, did, signature, status, trust_score, and EU AI Act compliance
- Passport is publicly verifiable and immutable, enabling audit trails
- Built to meet EU AI Act requirements for high‑risk AI agents
- Trust network records independent verifications, each signed and stored immutably
- API endpoints: /issue‑test, /verify/{agent_id}, /trust_network.html, /did/{agent_id}
- Open‑source SDK available on GitHub (raheem‑verisigil/verisigil‑ai)
- Planned features: behavioral fingerprinting, ZK compliance engine, MCP security scanner, federated trust network
- Provides a trust score and trust level based on verifier reputation