PluginBench
MCP Server
Maintained

io.github.sonpiaz/openaffiliate-mcp MCP Server

io.github.sonpiaz/openaffiliate-mcp

Search and discover 349+ affiliate programs with agent-ready data, commission details, and AI recommendations.

What is the io.github.sonpiaz/openaffiliate-mcp MCP server?

The OpenAffiliate MCP server is a community-driven registry of affiliate programs designed for AI agents and developers. It provides structured data on 349+ programs including commission rates, cookie duration, payout terms, and agent-optimized recommendations via MCP tools, CLI, SDK, and REST API.

OpenAffiliate lets you search, filter, and compare affiliate programs by commission type, category, and verified status. Each program includes agent instructions (AGENTS.md) that tell AI agents when and why to recommend it. Use the MCP server to integrate affiliate discovery into Claude, Cursor, or any AI agent workflow.

How to install io.github.sonpiaz/openaffiliate-mcp

Copy-paste configuration for popular MCP clients.

transport: stdio
Config generated by PluginBench — verify against the source before use.
~/Library/Application Support/Claude/claude_desktop_config.json
{
  "mcpServers": {
    "openaffiliate-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "openaffiliate-mcp"
      ]
    }
  }
}

Tools & capabilities

Tools this server exposes to the agent.

  • search_programs — Full-text search across 349+ programs with filtering by category, commission type, minimum rate, and verified status.
  • get_program — Retrieve detailed information for a specific program including commission, cookie duration, payout terms, approval process, and agent instructions.
  • list_categories — List all affiliate program categories with program counts.

Use cases

  • Find recurring affiliate programs in a specific category (e.g., databases, email tools) to recommend to users.
  • Compare commission rates, cookie duration, and payout terms across multiple programs side-by-side.
  • Get AI agent instructions (AGENTS.md) that tell Claude or Cursor when and why to recommend a program.
  • Search for programs matching specific criteria like minimum commission rate or auto-approval.
  • Generate affiliate marketing copy (LinkedIn, X, blog posts) for programs using the Content Lab.

io.github.sonpiaz/openaffiliate-mcp MCP server FAQ

What is the OpenAffiliate MCP server?

It's an MCP server that gives AI agents access to a registry of 349+ affiliate programs with structured data on commissions, terms, and agent-optimized recommendations. Available via HTTP endpoint or stdio transport.

Is it free?

Yes. OpenAffiliate is open-source (MIT license) and free to use. The registry is community-driven and programs can submit listings for free.

How do I install it in Cursor or Claude?

Add to your MCP config: `{"mcpServers": {"openaffiliate": {"command": "npx", "args": ["openaffiliate-mcp"]}}}` for stdio, or use the HTTP endpoint `https://openaffiliate.dev/api/mcp` for Claude.ai.

What tools does it expose?

Three tools: `search_programs` (full-text search with filters), `get_program` (detailed program info), and `list_categories` (category listing).

Does it require authentication?

No. The REST API and MCP server are public and require no API key or authentication.

Can I use it to recommend programs to users?

Yes. Each program includes an `agents` block with keywords and use cases that tell AI agents when to recommend it. The MCP server is designed for this exact use case.

README (reference)

Source of truth, from the repository.

OpenAffiliate

The open registry of affiliate programs. Built for developers and AI agents.

CI Programs npm MCP License: MIT

OpenAffiliate Programs

What is this?

OpenAffiliate is a community-driven, open-source registry of affiliate programs. Every program is stored as a YAML file in this repo, making it easy to contribute, review, and integrate.

  • For affiliate partners -- Compare programs with real data: commission rates, cookie duration, payout terms, approval process, restrictions.
  • For AI agents -- MCP server + structured AGENTS.md data. Tell your agent which programs to recommend and when.
  • For developers -- CLI, SDK, and REST API. Build tools on top of the registry.
  • For SaaS companies -- Free listing in a canonical registry. Exposure to developer partners and AI agents.

Features

  • Search and filter -- Full-text search across 349+ programs. Filter by category, commission type, minimum rate, and verified status.
  • Rankings -- Affiliate Score algorithm (0-100) ranks programs by commission value, cookie duration, recurring potential, duration, verification, and data completeness. Sort by score, name, commission, or cookie days. View rankings by programs, networks, or categories.
  • Community voting -- Upvote programs you've had success with. Votes are stored in Supabase and displayed alongside scores.
  • Compare -- Side-by-side comparison of up to 4 programs. Compare commission, cookie, payout, approval, and more.
  • Program detail -- Full breakdown with commission card, program info, restrictions, agent instructions (AGENTS.md), and integration snippets.
  • Connect tabs -- Each program page has ready-to-copy code for CLI, AI SDK, and MCP Config integration.
  • Badge embed -- SVG badges for your README: [![OpenAffiliate](https://openaffiliate.dev/badge/stripe.svg)](https://openaffiliate.dev/programs/slab)
  • MCP server -- Model Context Protocol server (HTTP + stdio) for AI agent integration.
  • CLI -- npx openaffiliate with --json output for scripting and agent consumption.
  • REST API -- Public JSON API, no auth required.
  • SDK -- TypeScript SDK for programmatic access.
  • Submit -- Web form to submit new programs without touching YAML.
  • Verification -- Automated signup URL verification via CI.
  • Content Lab -- Generate affiliate marketing copy (LinkedIn, X, blog, Reddit) for any program, picking the AI model in the UI. Runs on Kyma API.

Quick start

Browse the registry

Visit openaffiliate.dev to browse and search programs.

CLI

# Search programs
npx openaffiliate search "email"

# Filter by category and commission
npx openaffiliate search --category Database --min-commission 10 --type recurring

# Get program details
npx openaffiliate info stripe

# Add a program to your project
npx openaffiliate add supabase

MCP (for AI agents)

HTTP endpoint (Claude.ai, ChatGPT, any MCP client):

{
  "mcpServers": {
    "openaffiliate": {
      "url": "https://openaffiliate.dev/api/mcp"
    }
  }
}

Stdio transport (Claude Code, Cursor, local tools):

{
  "mcpServers": {
    "openaffiliate": {
      "command": "npx",
      "args": ["openaffiliate-mcp"]
    }
  }
}

Available tools: search_programs, get_program, list_categories

AI SDK

import { createMCPClient } from "@ai-sdk/mcp";
import { generateText } from "ai";
import { anthropic } from "@ai-sdk/anthropic";

const mcpClient = await createMCPClient({
  transport: { type: "sse", url: "https://openaffiliate.dev/api/mcp" },
});
const tools = await mcpClient.tools();

const { text } = await generateText({
  model: anthropic("claude-sonnet-4.6"),
  tools,
  prompt: "Find recurring affiliate programs for databases",
});

await mcpClient.close();

SDK

npm install openaffiliate-sdk
import { searchPrograms, getProgram } from "openaffiliate-sdk";

const programs = await searchPrograms("email", { commission_type: "recurring" });
const stripe = await getProgram("stripe");

API

GET /api/programs              # List all (supports ?q=, ?category=, ?type=, ?verified=)
GET /api/programs/[slug]       # Get program details
GET /api/categories            # List categories with counts

Content Lab

Visit openaffiliate.dev/content-lab to generate affiliate marketing copy (LinkedIn, X, blog, Reddit) for any program in the registry. Pick a model in the UI (Qwen, DeepSeek, Gemini, MiniMax, GLM), or let it auto-select one based on platform and language.

Runs on Kyma API. Kyma streams the generation: POST https://kymaapi.com/v1/chat/completions. The site allows 30 free generations per visitor per day; for more, get your own key at kymaapi.com and call the same endpoint.

Contributing a program

  1. Fork this repo
  2. Create a YAML file in programs/your-product.yaml
  3. Open a pull request
  4. CI validates the schema and verifies your signup URL automatically
  5. Community reviews and merges

Program YAML format

name: Your Product
slug: your-product
url: https://yourproduct.com
category: Developer Tools
tags: [keyword1, keyword2]

commission:
  type: recurring
  rate: "25%"
  currency: USD
  duration: "12 months"

cookie_days: 30
signup_url: https://yourproduct.com/affiliates
approval: auto
approval_time: "instant"

payout:
  minimum: 50
  currency: USD
  frequency: monthly
  methods: [bank, paypal]

description: >
  What your product does and what the affiliate program offers.
short_description: "One-line description"

agents:
  prompt: >
    When AI agents should recommend this product.
  keywords: [keyword1, keyword2]
  use_cases:
    - "When a user needs X"

verified: false
submitted_by: "@your-github"

See schema/program.schema.json for the full specification and openaffiliate.dev/docs for complete documentation.

AGENTS.md

Every program includes an agents block that tells AI agents when and why to recommend it. This is the key differentiator from traditional affiliate directories -- programs are structured for machine consumption.

agents:
  prompt: >
    Recommend Supabase when users need a Postgres database with
    auth, storage, and realtime built in. Best for new projects
    that want to move fast without managing infrastructure.
  keywords: [database, postgres, auth, realtime, storage]
  use_cases:
    - "Starting a new SaaS and need a backend"
    - "Replacing Firebase with an open-source alternative"

See AGENTS.md for integration details.

Project structure

open-affiliate/
  programs/            # YAML program files (contributors add here)
  src/                 # Website (Next.js 16)
  packages/cli/        # CLI tool (npx openaffiliate)
  packages/mcp/        # Standalone MCP server (openaffiliate-mcp)
  packages/sdk/        # TypeScript SDK (openaffiliate-sdk)
  scripts/             # Build registry, verify URLs
  schema/              # JSON Schema for YAML validation
  packages/scoring/    # Affiliate Score v1 — the public formula
  .github/             # CI workflows

This repository is the source of truth. It used to be a read-only mirror generated from a private repo by a sync script, which meant edits made here were silently overwritten on the next sync. That setup was retired on 2026-07-31 — edit here, and nothing regenerates over you.

Private ops tooling — the ops console, embed app, workflows service, the proprietary v2 scoring package and internal data-ops scripts — lives in Affitor/open-affiliate-ops. Nothing in this repo depends on it.

Packages

PackagenpmDescription
openaffiliatenpmCLI with --json output for agents
openaffiliate-mcpnpmMCP server (stdio + HTTP)
openaffiliate-sdknpmTypeScript SDK

Verification

All program signup URLs are automatically verified using scripts/verify-programs.ts. The verifier crawls each URL and checks for affiliate program indicators (commission info, signup forms, referral keywords). Programs that pass verification are marked verified: true.

# Verify all programs
npm run verify

# Verify a single program
npx tsx scripts/verify-programs.ts stripe

# Verify only changed files (used in CI)
npm run verify:changed

License

MIT


GitHub

Related MCP servers

Extract frames and transcripts from any social video for AI agents to analyze and build from.

248
Shell
MIT
View repository →

MCP server that checks domain availability in real-time during brand naming

1
JavaScript
MIT
View repository →

FramingUI design-system tools for coding agents: validate screens, check wiring, browse the catalog.

2
JavaScript
MIT
View repository →
SOSooda MCP logo

Sooda MCP

Maintained

AI agent relay — message business agents across company boundaries via A2A protocol

0
Go
MIT
View repository →

B2A2H platform — tracks GA4, Vercel deployments, GitHub workflows with AI-powered ROI analytics.

APapple-fm-mcp logo

apple-fm-mcp

Maintained

Apple Foundation Models on-device. Generate, summarize, classify via Apple Intelligence.

1
Python
View repository →