Patreon Mcp Server

io.github.KyuRishPublished by GitHub account KyuRishv0.1.1Updated Oct 24, 2026

Access your Patreon creator data - campaigns, patrons, tiers, and posts.

Overview

AI-generated overview

A local MCP server that gives an AI assistant read-only access to your own Patreon creator data: campaigns, tiers, patrons and posts.

What you can do
Read your authenticated Patreon profile and list your campaigns, with tier breakdowns for a single campaign. Browse patrons page by page, including pledge amounts, charge status and tier membership. Browse and read your posts page by page, or fetch a single post by ID.
When to use it
Useful when you want to ask an assistant questions about your Patreon business, such as top patrons by lifetime support, patron counts per tier, patrons with declining payments, or recent posts, without exporting data by hand.
Requirements
Runs locally as a Python process (Python 3.11+ and the uv package manager, per the README). Requires a Patreon Creator Access Token supplied in the PATREON_ACCESS_TOKEN environment variable. Network access to the Patreon API is needed.
Before you install
The PATREON_ACCESS_TOKEN is a secret credential and must be kept out of shared configs. Patron data such as names, pledge amounts and charge status is sent to your AI provider and may be retained under its policies; the README makes you responsible for complying with Patreon's Creator Privacy Promise and data protection laws. The server is read-only and stores no data itself.

Installation

In SourceWeft

  1. Open Patreon Mcp Server in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

Patreon MCP Server

[License: MIT] [Python 3.11+] [MCP] [PyPI]

mcp-name: io.github.KyuRish/patreon-mcp-server

Give AI assistants access to your Patreon creator data. The first authenticated Patreon MCP server - works with Claude Desktop, Cursor, Windsurf, VS Code Copilot, and any MCP-compatible client.

Quick Start

1. Get your Creator Access Token

Go to Patreon Developer Portal and copy your Creator Access Token. This token gives access to your own campaign data only.

2. Configure your MCP client

Claude Desktop - add to claude_desktop_config.json:

json
{  "mcpServers": {    "patreon": {      "command": "uv",      "args": ["run", "--directory", "/path/to/patreon-mcp-server", "src/patreon_mcp_server/server.py"],      "env": {        "PATREON_ACCESS_TOKEN": "your_token_here"      }    }  }}

Claude Code - add to .mcp.json in your project root:

json
{  "mcpServers": {    "patreon": {      "command": "uv",      "args": ["run", "--directory", "/path/to/patreon-mcp-server", "src/patreon_mcp_server/server.py"],      "env": {        "PATREON_ACCESS_TOKEN": "your_token_here"      }    }  }}

3. Start using it

Ask your AI assistant things like:

  • "Show me my Patreon campaigns"
  • "Who are my top patrons by lifetime support?"
  • "How many patrons are on each tier?"
  • "Which patrons have declining payments?"
  • "List my recent posts"

Available Tools

ToolDescriptionReturns
fetch_identityYour authenticated profileUser
fetch_campaignsList all your campaignsCampaign[]
fetch_campaignCampaign details with tier breakdownCampaignDetail
fetch_membersPaginated patron list (100/page)MemberPage
fetch_postsPaginated post list (20/page)PostPage
fetch_postSingle post by IDPost

Pagination: fetch_members and fetch_posts return a next_cursor field. Pass it as the cursor parameter to fetch the next page.

Data Fields

Member

full_name, patron_status, pledge_cadence, lifetime_support_cents, currently_entitled_amount_cents, last_charge_date, last_charge_status, will_pay_amount_cents, is_follower, tiers, user_name

Campaign

creation_name, patron_count, pledge_url, published_at, url, vanity, is_monthly, created_at, image_url, summary, one_liner, pay_per_name

Tier

title, amount_cents, description, published, patron_count

Post

title, content, is_paid, is_public, published_at, url, embed_data, embed_url

Privacy & Data

This server is designed with patron privacy in mind:

  • No patron emails - email addresses are never requested from the API
  • No private notes - creator notes about patrons are excluded
  • Read-only - no write operations, the server only reads your data
  • No data storage - the MCP server itself does not cache or persist any data

Important: When using this server with an AI assistant, patron data (names, pledge amounts, charge status) is sent to your AI provider (e.g., Anthropic, OpenAI) and may be temporarily retained per their data processing policies. You are responsible for ensuring your use complies with Patreon's Creator Privacy Promise and applicable data protection laws.

This project is not affiliated with or endorsed by Patreon.

Prerequisites

  • Python 3.11+
  • uv package manager
bash
# Clone the repogit clone https://github.com/kyurish/patreon-mcp-server.gitcd patreon-mcp-server
# Install dependenciesuv sync
# Test it runsPATREON_ACCESS_TOKEN=your_token uv run src/patreon_mcp_server/server.py

Project Structure

src/patreon_mcp_server/  server.py        # Entry point  mcp_server.py    # FastMCP init + client instance  tools.py         # @mcp.tool() definitions  models.py        # Pydantic models + JSON:API parsers  utils/    client.py      # PatreonClient (HTTP layer)

Roadmap

This server is currently read-only. Write operations (create posts, manage tiers, send messages to patrons) will be added if there's enough demand - open an issue or star the repo to show interest.

License

MIT License - see LICENSE for details.

Support

If you find this useful, consider supporting development on Patreon.

Source: README.md at commit 6cd1c54

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v0.1.1LatestSep 16, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts