T2
Building verifiable services
Issuance and verification on DIDComm and OpenID4VC, schemas and VTJSC, AnonCreds and SD-JWT, the MCP server.
In the tracks: Developer and Integrator, AI Agent Builder.
This module is an outline: the objectives, sections and assessment are agreed; the material is being written by the Showcase and Certification Working Group.
Outline. The material of this module is being written by the Showcase and Certification Working Group. The objectives, outline and assessment below are the agreed scope for program version 2026.1.
12 class hours. Assessed by: lab.
Learning objectives
After this module you can:
- Issue and verify credentials over DIDComm (AnonCreds) and over OpenID4VC (SD-JWT VC) from a VS Agent.
- Create a credential schema on the registry and use its VTJSC, type metadata and AnonCreds schema from the agent.
- Implement the trust decision on received presentations: branch on the verdict, never read claims otherwise.
- Verify peers before you connect and present your own credentials in flows.
- Expose a service's capabilities to AI agents through the Verana MCP server.
Prerequisites
C1, C2, C4, and T1 or an agent someone deployed for you. A programming language with an HTTP client.
Outline
1. The agent's model
Connections, credential records, presentation records, flows; events; idempotency; the two rails and when each is negotiated.
2. Schemas for builders
Authoring the JSON Schema; anchoring; what the registry publishes (VTJSC, type metadata, AnonCreds schema); referencing them from the agent.
3. Issuance
DIDComm issue-credential with AnonCreds; OpenID4VCI with SD-JWT VC; claim mapping and display metadata; revocation and status lists; sessions and fees.
4. Verification
DIDComm present-proof; OpenID4VP; what to request and what not to; the four verdicts and the evidence note; caching rules.
5. Verifying peers
The trust decision on connections; presenting your Service credential; vt-flow for onboarding evidence.
6. AI agents and MCP
The Verana MCP server; exposing a service's flows as tools; the agent's own identity (preview of T7).
Assessment
Lab: build a small service that issues one credential of a schema you created and verifies it back, on both rails, with the trust decision enforced. Graded on the test network against the rubric.
Reading list
docs.verana.io: Build, Verifiable services (issue, verify, trust decision, verify peers); Build, AI agents and MCP; Reference, Credentials; Reference, VS Agent Admin API. Verifiable Trust specification v4: Credentials; Trust resolution. VS Agent specification v4.