Trusty-Squire
MCP server that lets coding agents sign up for SaaS dev tools and manage stored
Trusty-Squire is an MCP server for agent tooling. It runs locally on your machine and is started by your AI client.
- Platforms
- CLI, API, macOS
- Runs
- Locally
- Version
- 1.0.33
- Listing updated
Similar MCP servers
Install
Run in a terminal
claude mcp add mcp -- npx -y @trusty-squire/mcpAdd to claude_desktop_config.json
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@trusty-squire/mcp"
]
}
}
}Add to ~/.cursor/mcp.json
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@trusty-squire/mcp"
]
}
}
}Add to .vscode/mcp.json
{
"servers": {
"mcp": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@trusty-squire/mcp"
]
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@trusty-squire/mcp"
]
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@trusty-squire/mcp"
]
}
}
}Using several servers? Build one config file with the MCP config generator.
About Trusty-Squire
The README says Trusty Squire is a credential broker for coding agents. Instead of pasting API keys into a chat, the agent asks it to sign up for a service, verify the email, and save the resulting key in a vault that the agent can use without reading the raw value.
It is installed through an npm package that runs a connect command, which signs the user in with Google or GitHub and registers the server with the chosen coding agent. The README lists Claude Code, Codex, Cursor, OpenCode, Goose, Cline, Continue, and Hermes as supported agents.
Beyond account creation, the README describes sign-in through a user's Google or GitHub account on sites that offer it, and purchases where the user approves once and the browser fills in card fields. Spending is bounded by a signed mandate the user approves. Developers who want approvals inside their own app can use the separate SDK.
What it does
- Automated API key signup
- The agent signs up for a service, completes email verification, and stores the returned key in a vault.
- Keys hidden from chat
- Saved keys are used by the agent without the raw value being shown in the conversation.
- Social sign-in
- Signs into sites through a user's Google or GitHub account where the site supports those providers.
- Approved purchases
- Card fields are filled in the browser after the user approves an order within a signed spending mandate.
- One-command install
- Setup runs through an npx connect command that registers the server with the chosen coding agent.
Good for
- Create a developer account and get an API key for a SaaS tool without leaving the coding session
- Store third-party API keys in a vault so agents can call APIs without exposing secrets
- Let an agent sign in to a site with an existing Google or GitHub account
- Place an online order from a coding agent with a one-time approval
Trusty-Squire FAQ
How is the server installed?
The README gives the command npx -y @trusty-squire/mcp@latest connect, which signs the user in with Google or GitHub and adds the server to the chosen coding agent.
Which coding agents are supported?
The README lists Claude Code, Codex, Cursor, OpenCode, Goose, Cline, Continue, and Hermes. A specific agent can be chosen with the --target option.
Does the agent see stored API keys?
The README says the agent uses a saved key without ever seeing it, and that new keys are stored in the vault instead of the chat.
How do I add Trusty-Squire to Claude Code?
Run this command in your terminal: claude mcp add mcp -- npx -y @trusty-squire/mcp
What are the best alternatives to Trusty-Squire?
The closest alternatives to Trusty-Squire are Trust, vivioo.io, azeth-protocol, ArcAgent MCP and 123skills.site.
Where this comes from
- Description: written from the project's README and its MCP Registry entry.
- Install steps: from the official MCP Registry entry.