Paxmod MCP server

Connect Claude, Cursor, VS Code or any other MCP client to Paxmod. Your agent can read the API guide, get a test key with no signup, check messages, and test a moderation policy against your own examples before you save it.

Hosted at https://www.paxmod.com/mcp (Streamable HTTP). There is nothing to install and no sign-in: tools that work on a project take its API key.

Connect your agent

Claude Code

claude mcp add --transport http paxmod https://www.paxmod.com/mcp

Claude and Claude Desktop

Open Settings, then Connectors, choose Add custom connector and enter https://www.paxmod.com/mcp.

Cursor

Add this to .cursor/mcp.json in your project, or ~/.cursor/mcp.json for every project.

{
  "mcpServers": {
    "paxmod": { "url": "https://www.paxmod.com/mcp" }
  }
}

VS Code

Add this to .vscode/mcp.json.

{
  "servers": {
    "paxmod": { "type": "http", "url": "https://www.paxmod.com/mcp" }
  }
}

Other clients

Add a remote MCP server with the URL above and the Streamable HTTP transport.

Tools

Tool	What it does	Needs
get_integration_guide	The full API guide for coding agents: endpoints, request and response shapes, errors, rate limits and code samples for Node, Python, Unity, Roblox and more.	Nothing
create_test_api_key	A working text moderation key with no signup: 100 messages over 72 hours. Returns a claim link so the developer can keep the key.	Nothing
moderate_text	Checks one message with the project's settings and returns flagged or not, with a score per category.	API key
test_policy	Runs up to 50 example messages against a draft or saved Advanced mode policy and says which come back as expected. Nothing is stored and no credits are used, except on a free test key, where each message checked counts toward its 100.	API key
test_word_rules	Shows what word rules do to each message: block, skip moderation, or rewrite before checking. Draft or saved rules, no model call.	API key

Tune your policy with your agent

Advanced mode moderates against a policy you write in plain English. When it flags something it shouldn't, or misses something, give your agent the examples and ask it to fix the policy:

Our Paxmod policy is flagging some messages it shouldn't. These 6 were flagged
but are fine: ... These 2 got through and shouldn't have: ...
Rewrite the policy so all 8 come back right, and show me the new policy.

The agent rewrites the policy and calls test_policy with the draft until every example comes back right. Your word rules run first, as they do live. Tests store nothing and use no credits on your project (on a free test key, each message checked counts toward its 100). When you're happy, paste the policy into your project's Settings in the console. The same test is on the Settings page as Test your policy, and over the API as POST /api/v1/test-policy.

Adding Paxmod to your app

Ask your coding agent to "add Paxmod moderation to our chat". It reads the guide, creates a test key, builds the integration and checks it with real messages, then gives you the claim link to keep the key. Agents without MCP can start from SKILL.md.

API reference Pricing