Works in Claude · ChatGPT · Cursor · Windsurf

Your GTM pipeline,
in plain language.

Connect RichAPI to Claude or ChatGPT once. Then just ask: "Find 20 VPs of Marketing at Series-B SaaS companies who posted about pipeline this week, with verified emails." RichAPI does the data work behind the scenes — searching, enriching, verifying across 50+ sources — and your assistant hands back the list with a cost receipt. No code. No CSV shuffling. No platform subscription.

Claude@richapi · 35+ tools
Find 20 Series-B SaaS CTOs who posted about hiring this week, get verified emails
richapi_post_keyword_search20 authors · 6 cr
richapi_enrich_profile×20 · 20 cr
richapi_email_finder14 hits · 6 soft misses
✓ 20 profiles · 14 verified emails · 60 credits ($1.20) · misses billed at half rate6 search + 20 enrich + 28 hits + 6 soft misses = 60

RichAPI MCP is a Model Context Protocol server for GTM data. It gives Claude, Cursor, Windsurf, and ChatGPT 35+ tools — LinkedIn profile/company enrichment, people search, post and hiring signals, and five multi-provider waterfalls (email finding, phone finding, email verification, person and company enrichment) — from the same key and credit pool as the REST API at api.richapi.ai. Billing is success-weighted: full price only on results.

What you can ask for

If you can describe your buyer, you can build the list.

The same actions are available as a REST API for whoever on your team writes code — same account, same credits.

You stay in control

Built so you can let an assistant loose on it.

Connect

Connected in about a minute.

Three plain steps
  1. Create your account app.richapi.ai/signup (25 free credits, no card)
  2. Connect your assistant → In Claude: Settings → Connectors → RichAPI → Connect. (This path goes live with the Anthropic directory listing; until then, the config route below works today.)
  3. Run your first starter prompt → pick one from the library below; your first verified list arrives with a receipt.
For implementation owners — manual config (Claude Desktop, Cursor, Windsurf, ChatGPT)
Claude Desktop / Cursor — via mcp-remote shim
{ "mcpServers": { "richapi": {
    "command": "npx",
    "args": ["mcp-remote", "https://mcp.richapi.ai/mcp",
             "--header", "x-api-key: ${RICHAPI_KEY}"] } } }
Windsurf — native remote
{ "mcpServers": { "richapi": {
    "serverUrl": "https://mcp.richapi.ai/mcp",
    "headers": { "x-api-key": "${RICHAPI_KEY}" } } }

Per-client guides (Continue, Zed, VS Code, ChatGPT connectors): docs.richapi.ai/mcp — client config formats change faster than marketing pages; the docs page is canonical.

Starter prompt library

Copy, paste, edit the brackets, run.

Each prompt shows its receipt when it finishes. Typical cost of the weekly pipeline fill: ~60 credits ≈ $1.20.

Why one connector beats five

Every data vendor ships an MCP server for their own database now. Connect five and your assistant juggles five logins, five bills, and five tool lists. RichAPI is one connector whose tools run across providers automatically — one schema, one invoice, billed by outcome. Fewer tools is also how assistants stay accurate: less catalog, better choices.

Pricing

Same bundles as the REST API

Same credit pool whether your assistant calls it or your code does. Full price only on results. See full pricing →

FAQ

Frequently asked questions

Your next list is one prompt away.