Dock Labs today announced the launch of its Model Context Protocol (MCP) integration, a new capability designed to bring secure digital ID credential operations into AI and agent-driven workflows.
The MCP server enables organizations to issue and verify credentials, manage DIDs, and generate presentation requests directly through LLM-powered agents, while maintaining tight control over what those agents are allowed to do.
Purpose-built for the emerging agentic commerce ecosystem, the integration acts as a secure intermediary between autonomous systems and Dock Labs’ digital ID infrastructure. Rather than granting agents broad API access, the MCP layer allows enterprises to expose tightly scoped identity functions, supporting least-privilege access and safer automation.
As AI agents begin to participate in real transactions and cross-system workflows, Dock Labs is positioning verifiable identity as foundational infrastructure for the next generation of trusted, autonomous digital interactions.
The Problem: Agents Don’t Natively Understand Digital ID
As agents become more autonomous, they will inevitably need to answer questions like:
- Who am I acting on behalf of?
- Is the counterparty trustworthy?
- Do I have permission to complete this transaction?
In theory, you could teach an agent how to work directly with digital ID infrastructure APIs. In practice, that approach quickly breaks down.
Giving an agent full API access creates a new class of risk. Agents are non-deterministic by nature, and unrestricted access makes it difficult to enforce least privilege or predictable behavior.
What’s needed is a structured control layer between agents and identity infrastructure.
This is exactly where MCP comes in.
What MCP Is
Model Context Protocol (MCP) provides a standardized way for agents and LLMs to safely call external tools and services.
Instead of hard-wiring digital ID credential logic into an agent, MCP allows you to:
- Expose specific capabilities as tools
- Tightly scope what the agent can do
- Maintain control over sensitive operations
Think of MCP as the safe adapter layer that lets agents expand their capabilities without giving them the keys to the kingdom.
What We Built
Our new MCP server wraps our existing digital ID infrastructure and makes it callable directly by agents and LLMs using natural language workflows.
In simple terms: we’ve made digital ID operations agent-accessible.
Conversational Digital ID Operations
With the MCP integration, agents (or LLM interfaces) can now:
- Create new Decentralized Identifiers (DIDs)
- Configure DID profiles (name, logo, metadata)
- Issue digital ID credentials
- Create proof requests
- Verify credentials
The experience is conversational.
For example, an LLM can:
- Ask for the schema
- Prompt for required attributes
- Issue the credential
- Generate a verification request
Under the hood, the MCP server is wrapping our REST API and exposing it in a form agents can safely use.
Built for the Agentic Future
It’s important to be clear: this release is less about humans issuing credentials in chat windows, and more about preparing infrastructure for autonomous systems.
The real value emerges when agents begin interacting with other agents, enterprise systems and consumer identity flows.
We expect growing demand around:
- Agent-to-agent trust
- Delegated authority
- Autonomous transaction approval
- Verifiable machine identity
In these environments, agents must be able to prove and verify authority programmatically, and do so safely.
This MCP integration is an early but important building block toward that future.
Example: Agent Delegation
Imagine a travel planning agent that handles the user experience but delegates hotel booking to the hotel agent.
Before completing the handoff, the agents may need to:
- Verify authority
- Exchange proofs
- Delegate scoped permissions
Our MCP layer enables those digital ID operations without requiring the agent itself to understand the full complexity of verifiable credentials.
Why MCP Instead of Direct API Access?
This is where the security story becomes critical.
Without MCP, the typical approach would be to give the agent API keys. That introduces several risks:
- Unrestricted access
- Difficult policy enforcement
- Unpredictable agent behavior
- Expanded blast radius if something goes wrong
With the MCP server in place, organizations can tightly control what the agent is allowed to do.
For example, you can allow an agent to verify credentials, issue only specific credential types or use predefined proof templates.
While preventing it from creating new schemas, generating new DIDs or performing sensitive admin operations.
This enables a least-privilege model for AI agents, which will become increasingly important as autonomous systems scale.
Full Control
The MCP server is designed for enterprise control and flexibility.
In a typical deployment, the MCP server runs within your own environment. It is designed to sit behind your firewall, giving your team full control over how and where credential operations are executed.
Once deployed, the MCP server is simply configured as a tool available to the agent. The integration itself is lightweight and straightforward to wire up within the agent configuration, allowing teams to extend agent capabilities without introducing unnecessary complexity or external dependencies.
This avoids the need for a centralized hosted dependency and aligns with enterprise security expectations.
What’s Coming Next
This is just the first step.
Work is already underway on a second MCP server focused on wallet functionality, including:
- Holding credentials
- Presenting credentials
- More specialized agent flows
The goal is not to create one overly powerful tool, but a set of purpose-built, composable identity capabilities for agents.
How To Use The Truvera MCP Server
The MCP server is currently in a private repo.
Planned availability includes:
- Public repository access
- The Dock Labs Non-Production licensing model, similar to our Wallet SDK
- Docker container distribution
- Simple configuration using API keys
This will allow builders and partners to experiment quickly in their own environments.
Get in touch with the Dock Labs team to request early access.
The Bigger Picture
AI agents are rapidly becoming economic actors.
As that happens, identity and authority can no longer remain human-only concerns. They must become programmable, verifiable, and safely consumable by autonomous systems.
MCP is emerging as one of the key connective layers, and digital ID credentials are a natural trust primitive for that world.
This launch is an early step, but an important one.
As agents become primary actors in digital systems, the ability to verify identity and authority programmatically will move from nice-to-have to mandatory infrastructure.
We’re building toward that future, and this is just the beginning.
