MCP
Model Context Protocol (MCP) enables AI tools and agents to seamlessly integrate with Trustless Work’s escrow infrastructure through a unified and context-aware interface.
Last updated
Model Context Protocol (MCP) enables AI tools and agents to seamlessly integrate with Trustless Work’s escrow infrastructure through a unified and context-aware interface.
Use MCP to let Cursor read Trustless Work docs and trigger escrow workflows from chat.
Use both servers together:
trustlesswork-docs gives Cursor direct access to the docs.
trustlesswork gives Cursor access to Trustless Work tools and actions.
Use the docs server for answers and code generation.
Use the product server for escrow actions and live operations.


Open a new Cursor chat.
Switch to Agent Mode.
Then ask Cursor to work with Trustless Work directly.
Example prompts:
Create a new multi-release escrow with the SDK.
Generate code to call the changeMilestoneStatus endpoint.
Show me how to sign a transaction for releaseFunds.
For SDK actions, install the Trustless Work SDK and wallet dependencies in your project first.
Ask Trustless Work questions in plain language
Generate SDK integrations faster
Walk through XDR signing flows
Trigger escrow operations from your editor
This gives you an AI-native workflow for Trustless Work development.
If the server does not connect, reload Cursor before changing the config again.
Make sure mcp.json is in your project root.
Check these values:
the URL has no trailing slash
type is streamable-http
your network allows outbound requests
Check these basics:
Agent Mode is enabled
you are in the right workspace
the JSON file has no syntax errors
Last updated
{
"mcpServers": {
"trustlesswork-docs": {
"type": "streamable-http",
"url": "https://docs.trustlesswork.com/trustless-work/~gitbook/mcp",
"headers": {}
},
"trustlesswork": {
"type": "streamable-http",
"url": "https://mcp.trustlesswork.com/mcp",
"headers": {}
}
}
}