Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.influship.com/llms.txt

Use this file to discover all available pages before exploring further.

The Influship MCP server exposes the API to AI agents — Claude Desktop, Cursor, ChatGPT Connectors, VS Code Copilot, and any Model Context Protocol-compatible client — as eight typed tools the agent can call directly. The server is hosted at https://mcp.influship.com/mcp and uses the Streamable HTTP transport (spec 2025-06-18). Stdio-only clients connect via the mcp-remote bridge.

​
How it works

Your existing API key, billing, and rate limits all apply — no separate setup, no separate billing.
  1. The agent connects to https://mcp.influship.com/mcp with your X-API-Key
  2. The agent discovers eight typed tools, each with a full description and parameter schema
  3. The agent picks the right tool (e.g. search_creators) and calls it with its arguments
  4. The result comes back in the same format as the REST API
Clients can list available tools without an API key, so users can browse the tool surface before entering credentials. Running a tool requires X-API-Key.

​
Tools

ToolMethodEndpoint
search_creatorsPOST/v1/search
autocomplete_creatorsGET/v1/creators/autocomplete
find_lookalike_creatorsPOST/v1/creators/lookalike
match_creatorsPOST/v1/creators/match
get_creatorGET/v1/creators/:id
get_profileGET/v1/profiles/:platform/:username
lookup_profilesPOST/v1/profiles/lookup
get_postsGET/v1/posts
Each tool’s description in the MCP server tells the agent when to pick it over the others (e.g. “use find_lookalike_creators when you have seed creator IDs; use search_creators when you only have a topic”). The agent uses these descriptions to route requests to the right tool.

​
Quickstart

Get an API key from the dashboard, then add the server to your client.

​
Claude Code (CLI)

One command: 
claude mcp add influship --transport http https://mcp.influship.com/mcp --header "X-API-Key: YOUR_KEY"
Add --scope user to install across every project, not just the current directory.

​
Claude Desktop

Settings → Connectors → Add Custom Connector. Set the URL to https://mcp.influship.com/mcp and the header X-API-Key: <your_key>. For older Claude Desktop versions or if you prefer a config file at ~/Library/Application Support/Claude/claude_desktop_config.json: 
{
  "mcpServers": {
    "influship": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://mcp.influship.com/mcp",
        "--header",
        "X-API-Key:YOUR_KEY"
      ]
    }
  }
}

​
Cursor / Codex / VS Code

Settings → Tools & MCP → New MCP Server: 
{
  "mcpServers": {
    "influship": {
      "type": "http",
      "url": "https://mcp.influship.com/mcp",
      "headers": {
        "X-API-Key": "YOUR_KEY"
      }
    }
  }
}

​
Windsurf

Settings → Tools → MCP → Add server. Same JSON shape as Cursor: 
{
  "mcpServers": {
    "influship": {
      "type": "http",
      "url": "https://mcp.influship.com/mcp",
      "headers": {
        "X-API-Key": "YOUR_KEY"
      }
    }
  }
}
If your Windsurf version doesn’t support HTTP MCP servers yet, use the stdio bridge below.

​
ChatGPT (Apps & Connectors)

Use the custom MCP connector flow. URL https://mcp.influship.com/mcp, auth type “API key”, header name X-API-Key.

​
Stdio-only clients

Some clients only support stdio transport. Bridge them with mcp-remote: 
npx -y mcp-remote https://mcp.influship.com/mcp --header "X-API-Key:YOUR_KEY"

​
Authentication and billing

The MCP server uses your existing API key as the authentication and billing surface. There is no separate “MCP plan” — every call passes through to the regular API:
  • The X-API-Key you set in the MCP client is forwarded to api.influship.com on every tool call
  • Your subscription tier, rate limits, and usage caps apply exactly as they do via the REST API
  • Invoices come through the same Stripe customer record
  • Calls show up in the same usage logs in the dashboard
If you don’t have an API key yet, grab one — there’s a free tier you can try the MCP server on.

​
Pricing

MCP tool calls bill at the same per-credit price as the underlying REST endpoint. There’s no MCP surcharge.
ToolCost
search_creators25 credits + 2 per result returned
find_lookalike_creators1.5 per creator returned
match_creators1 per creator scored
autocomplete_creators0.1 per call
get_creator0.1 per call
get_profile0.1 per call
lookup_profiles0.1 per profile in the request
get_posts0.05 per post returned
One credit is $0.01. See the REST pricing reference for the full breakdown and per-tier discounts.

​
Rate limits

Per-API-key limits apply (same as the REST API). There’s also a per-IP cap of 60 requests per minute — if you hit it you’ll get a 429 with a retry-after header. If you’re connecting many agents from the same IP (a corporate proxy, for example), reach out and we’ll tune the IP cap.

​
Error codes

When a tool call fails, the error message is surfaced as readable text in the MCP client.
ErrorMeaning
400Missing or invalid parameter
401Missing or invalid X-API-Key
402Payment required
403Key valid but lacks permission for this tool
404Creator, profile, or post not found
429Rate limit exceeded — check retry_after_seconds in the response
5xxServer error

​
Anonymous agents (no API key)

For one-off agents that don’t have an API key — e.g. an autonomous research agent paying per-call — use the direct HTTP API with x402 or MPP. Same endpoints, same data, pay-per-request via USDC or Stripe cards. No signup required. MCP-native MPP support (where the agent pays the MCP server directly per tool call) is planned. Until then, the MCP server is API-key only.

​
Implementation tips

  • Trust tool descriptions. Each tool’s description is written for the LLM. It explains when to pick that tool over the others. Don’t override or rewrite it client-side.
  • Pass real IDs. Tools that take a creator_id expect a UUID from a previous tool’s response. Resolving by platform + username is supported on lookalike, match, and posts — useful when the user types a handle.
  • Keep limit honest. Larger limit on search_creators and find_lookalike_creators costs more and usually surfaces lower-quality results past the top 15–20. Start small.
  • Cache get_creator and get_profile. Both are cheap, and creator/profile data doesn’t change often. If your agent runs many sessions for the same user, cache.

​
Next steps

Get an API Key

Sign up and grab your key from the dashboard

API Reference

Full endpoint specs and response schemas — the tools mirror these

Pay with x402

Anonymous per-request payment for one-off agents

Pay with MPP

Stripe cards or Tempo USDC for agent-native payments