AgentAuth MCP Server
MCP (Model Context Protocol) server that exposes AgentAuth identity and authentication tools to any MCP-compatible AI agent.
What it does
Instead of writing custom HTTP client code, any MCP-compatible agent (Claude, OpenAI Assistants, LangChain agents, etc.) can use AgentAuth through standard MCP tool calls:
- discover — Learn what the AgentAuth service offers
- quickstart — Register a new agent and get credentials in one call
- authenticate — Exchange an API key for an access token
- refresh_token — Refresh an expiring token
- introspect_token — Check if a token is valid
- revoke_token — Invalidate a token
- create_credential / rotate_credential / revoke_credential — API key lifecycle
- create_delegation — Delegate permissions to another agent
- check_permission — Pre-flight authorization check
- list_agents / get_agent — Agent discovery
Quick Start
The MCP endpoint is built into the AgentAuth API — no separate server or local install required.
Add this to your MCP client config:
{
"mcpServers": {
"agentauth": {
"url": "https://agentauth.radi.pro/mcp"
}
}
}
That's it. Every MCP-compatible client can connect with just a URL.
Client Configuration
Claude Desktop
Edit ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):
{
"mcpServers": {
"agentauth": {
"url": "https://agentauth.radi.pro/mcp"
}
}
}
Restart Claude Desktop after saving.
Claude Code (CLI)
Add to .claude/settings.json in your project root, or globally at ~/.claude/settings.json:
{
"mcpServers": {
"agentauth": {
"url": "https://agentauth.radi.pro/mcp"
}
}
}
Cursor
Edit ~/.cursor/mcp.json (global) or .cursor/mcp.json (project-scoped):
{
"mcpServers": {
"agentauth": {
"url": "https://agentauth.radi.pro/mcp"
}
}
}
GitHub Copilot (VS Code)
Add to VS Code settings.json (Ctrl+Shift+P → Open User Settings (JSON)):
{
"github.copilot.chat.mcpServers": {
"agentauth": {
"url": "https://agentauth.radi.pro/mcp"
}
}
}
For workspace-scoped config, add the same block to .vscode/settings.json.
Windsurf (Codeium)
Edit ~/.codeium/windsurf/mcp_config.json:
{
"mcpServers": {
"agentauth": {
"url": "https://agentauth.radi.pro/mcp"
}
}
}
Continue.dev
Edit ~/.continue/config.json:
{
"mcpServers": [
{
"name": "agentauth",
"url": "https://agentauth.radi.pro/mcp"
}
]
}
Zed
Edit ~/.config/zed/settings.json:
{
"context_servers": {
"agentauth": {
"url": "https://agentauth.radi.pro/mcp"
}
}
}
OpenHands (formerly OpenDevin)
In config.toml:
[[mcp_servers]]
name = "agentauth"
url = "https://agentauth.radi.pro/mcp"
Any MCP-compatible agent (Python SDK)
from mcp import ClientSession
from mcp.client.streamable_http import streamablehttp_client
async with streamablehttp_client("https://agentauth.radi.pro/mcp") as (read, write, _):
async with ClientSession(read, write) as session:
await session.initialize()
tools = await session.list_tools()
Self-hosting
If you run your own AgentAuth instance, the MCP endpoint is automatically available at /mcp. No extra configuration needed — it's part of the main application.
https://your-agentauth-instance.com/mcp
The MCP server internally calls the AgentAuth REST API via AGENTAUTH_URL. When running as part of the main app, set this to the internal URL:
AGENTAUTH_URL=http://127.0.0.1:8000
Local (stdio) mode
For development or air-gapped environments, you can also run the MCP server locally as a stdio subprocess. This requires the package installed on the client machine.
cd mcp-server
uv pip install -e .
Then configure your MCP client with a command instead of a URL:
{
"mcpServers": {
"agentauth": {
"command": "uv",
"args": ["run", "--directory", "/path/to/agent-auth/mcp-server", "agentauth-mcp"],
"env": {
"AGENTAUTH_URL": "https://agentauth.radi.pro"
}
}
}
}
Usage flow
A typical agent interaction looks like:
- First time — call
quickstartto register and get credentials - Subsequent sessions — call
authenticatewith saved API key to get a token - During work — use the token for
list_agents,create_delegation,check_permission, etc. - Before token expires — call
refresh_token(checkrefresh_beforefrom authenticate response) - When done — optionally
revoke_token
Environment variables
| Variable | Required | Description |
|---|---|---|
AGENTAUTH_URL |
Yes | Base URL of the AgentAuth service (e.g. https://agentauth.radi.pro) |
AGENTAUTH_API_KEY |
No | Default API key for the authenticate tool |
Development
# Install dev dependencies
uv pip install -e ".[dev]"
# Run directly (stdio)
AGENTAUTH_URL=http://localhost:8000 python -m agentauth_mcp.server
Metadata
Release files for agentauth-mcp-server 0.1.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| agentauth_mcp_server-0.1.1.tar.gz | 57.8 kB | Without provenance | Details for agentauth_mcp_server-0.1.1.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| agentauth_mcp_server-0.1.1-py3-none-any.whl | Python 3 | none | any | Details for agentauth_mcp_server-0.1.1-py3-none-any.whl |
Total release size: 66.9 kB
Release files / agentauth_mcp_server-0.1.1.tar.gz
| Download URL | agentauth_mcp_server-0.1.1.tar.gz |
|---|---|
| Size | 57.8 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
933d8b54ef07006d0ea7df2bdb59562557bca13dd726c2480199baf0991955ef
|
|
BLAKE2b-256 checksum How to use checksums |
ad82f5c891f6452da4a3cc36effef3acc9e39ae54c02630ba4ede38916b1add7
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.9.26 {"installer":{"name":"uv","version":"0.9.26","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}
|
Release files / agentauth_mcp_server-0.1.1-py3-none-any.whl
| Download URL | agentauth_mcp_server-0.1.1-py3-none-any.whl |
|---|---|
| Size | 9.2 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
e709f7fa419ec058da0c6087bacc5e65b6903ec71e71a450b4258652d070800f
|
|
BLAKE2b-256 checksum How to use checksums |
34cb738d6e9a54fa235d4e4c7f548c4f1c26787878beded644cf8bbae6743373
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.9.26 {"installer":{"name":"uv","version":"0.9.26","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}
|
Security for agentauth-mcp-server 0.1.1
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.