About agentfuse-mcp
agentfuse-mcp connects MCP-compatible AI agents to the AgentFuse affiliate API without requiring them to write HTTP code. It supports browsing programs, creating tracked links, and logging conversions.
Key features
- 01Browse affiliate programs by category
- 02Get program details by slug
- 03Generate tracked attribution links
- 04View clicks, signups, and commission stats
- 05Record signups and commissions
- 06List tracked links
Use cases
- 01Find affiliate programs in a category
- 02Create tracked links for users
- 03Review clicks, signups, and commissions
- 04Log signup or purchase conversions
MCP support
The MCP server exposes AgentFuse affiliate API operations as tools for MCP-compatible AI agents, including program browsing, tracked-link creation, statistics, and conversion recording.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
npx -y agentfuse-mcpAGENTFUSE_API_KEYSecretYour AgentFuse API key (get one at agentfuse.io). Format: af_live_...AGENTFUSE_API_URLAgentFuse API base URL. Defaults to https://api.agentfuse.ioLocal servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.
Tools
7Saved tool list · updated 3w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| limit | number | No | Number of programs to return (default: 50, max: 100). |
| cursor | string | No | Pagination cursor returned from a previous call. Pass this to get the next page. |
| category | string | No | Filter by category slug. Examples: productivity, marketing, ai-voice, ai-writing, email-marketing, newsletter, automation, marketing-automation, seo-analytics, design, developer-tools. Omit to return all categories. |
Details
From the project READMEagentfuse-mcp
MCP server for AgentFuse -- the affiliate API middleware for AI agents.
Lets any MCP-compatible AI agent (Claude Code, Cowork, custom agents) browse affiliate programs, generate tracked links, and record conversions without writing HTTP code.
Tools
| Tool | Description |
|---|---|
list_affiliate_programs | Browse all programs in the catalog (filterable by category) |
get_affiliate_program | Get full details for one program by slug |
generate_tracked_link | Create an attribution-tagged short link for a user |
get_stats | Dashboard summary: clicks, signups, commissions |
record_signup | Log a conversion event when a user signs up |
list_tracked_links | List all tracked links generated for your users |
record_commission | Log a commission event (purchase or renewal) |
Requirements
- Node.js >= 18
- An AgentFuse API key -- get one at agentfuse.io
Usage
Option 1 -- Claude Code (recommended)
Add to your project's .claude/mcp.json (or your global ~/.claude/mcp.json):
{
"mcpServers": {
"agentfuse": {
"command": "npx",
"args": ["-y", "agentfuse-mcp"],
"env": {
"AGENTFUSE_API_KEY": "af_live_your_key_here"
}
}
}
}
Or, if you've cloned this repo locally:
{
"mcpServers": {
"agentfuse": {
"command": "node",
"args": ["/path/to/agentfuse-mcp/src/index.js"],
"env": {
"AGENTFUSE_API_KEY": "af_live_your_key_here"
}
}
}
}
Option 2 -- Run directly
npm install
AGENTFUSE_API_KEY=af_live_... node src/index.js
Option 3 -- Test the tool list (no API key needed)
node src/index.js --test
Example agent interactions
Browse programs:
"List all affiliate programs in the AI tools category" -> calls
list_affiliate_programs({ category: "ai-voice" })
Generate a link:
"Create a tracked Webflow affiliate link for user [email protected]" -> calls
generate_tracked_link({ program_slug: "webflow", end_user_id: "[email protected]" })
Check stats:
"How many clicks and signups have we had this month?" -> calls
get_stats()
Environment variables
| Variable | Required | Default | Description |
|---|---|---|---|
AGENTFUSE_API_KEY | Yes | -- | Your AgentFuse API key (af_live_...) |
AGENTFUSE_API_URL | No | https://api.agentfuse.io | Override the API base URL |
License
MIT (c) AgentFuse / GSK AIOrch LLC
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.