The MolTrust MCP Server is now listed on Glama.ai, one of the largest directories for Model Context Protocol servers with over 1,000 indexed servers. Every tool has been indexed and is discoverable by any MCP-compatible client.

8 Tools
Identity, reputation, credentials, credits, and on-chain agent resolution — all available as standard MCP tool calls

Why MCP Matters for Agent Trust

The Model Context Protocol, developed by Anthropic, standardizes how LLMs interact with external tools. An LLM running behind Claude Desktop, Cursor, or any MCP client can call tools exposed by MCP servers — without custom integration code.

For agent trust, this is significant. Before MCP, checking an agent's identity or reputation required custom API calls, authentication handling, and response parsing. With MCP, any LLM can verify an agent's identity in a single tool call:

// Claude asks: "Is this agent trustworthy?" moltrust_verify("did:moltrust:a1b2c3d4e5f60718") // Returns: verified identity, trust score, registration date

No SDK. No API keys for read operations. No boilerplate. The LLM decides when to check trust — and it has the tools to do so.

The 8 Tools

moltrust_register
Register a new agent identity. Returns a W3C DID and a signed Verifiable Credential.
moltrust_verify
Verify any agent's DID. Returns identity status, trust score, platform, and registration date.
moltrust_reputation
Query an agent's reputation score (1–5) and total number of ratings.
moltrust_rate
Rate an agent after an interaction. Builds the decentralized trust graph.
moltrust_credential
Issue or verify W3C Verifiable Credentials. Ed25519-signed, tamper-proof.
moltrust_credits
Check credit balance and API pricing for a given DID.
moltrust_deposit
Generate a USDC deposit address on Base for credit top-ups.
moltrust_erc8004
Query on-chain ERC-8004 Agent Cards and resolve agents by their on-chain ID.

ERC-8004: The On-Chain Dimension

The moltrust_erc8004 tool is new in v0.4.0 and connects MCP clients to on-chain agent identities on Base. ERC-8004 ("Trustless Agents") defines three on-chain registries — Identity, Reputation, and Validation — deployed across 16 chains.

MolTrust bridges both worlds. When you register an agent through MolTrust, it gets a W3C DID for off-chain interactions and an ERC-8004 Agent Card on Base. One identity, two layers of trust.

Through MCP, any LLM can now:

Installation

Install from PyPI:

pip install moltrust-mcp-server

Or run directly with uvx:

uvx moltrust-mcp-server

For Claude Desktop, add this to your claude_desktop_config.json:

{ "mcpServers": { "moltrust": { "command": "uvx", "args": ["moltrust-mcp-server"], "env": { "MOLTRUST_API_KEY": "your_key_here" } } } }

Read-only operations (verify, reputation, ERC-8004 lookups) work without an API key. Registration, rating, and credential issuance require a free key from moltrust.ch.

Where to Find Us

Try the MCP Server

Install with pip install moltrust-mcp-server and give your LLM the ability to verify agent identities, check reputations, and query on-chain trust data.

View on GitHub →

MolTrust is a trust infrastructure platform for AI agents. We provide identity verification, reputation scoring, and W3C Verifiable Credentials — bridged to ERC-8004 on Base. Learn more at moltrust.ch.