find_contactResolve a person to verified public profiles.

Your agent has a name and a company. It needs the right X handle, the right LinkedIn URL, the right Substack, and the right public email — fast, and without falling for impersonators. find_contact resolves a person across public sources, scores each candidate match for confidence, and returns the verified set with their most recent activity. Two credits. About two seconds.

When to call this tool

This is the right tool whenever the agent has a name and needs the canonical public-profile set before doing anything outbound. The tool returns only matches that pass a confidence threshold, and is honest about ambiguity — when there are multiple plausible "John Smith"s at the same company, the response includes a candidates array with disambiguating signal so the agent can ask the user or score against context.

It does not return private email addresses, phone numbers, or any signal scraped from regulated data brokers. It returns the public addressable surface — handles, public newsletters, public blogs, public talks, recent posts. That is precisely what an outbound agent needs to write a thoughtful first message.

Input schema

{
  "name": "string  // required",
  "company": "string  // optional",
  "role": "string  // optional",
  "min_confidence": "number  // 0-1, default 0.7"
}

Output schema

{
  "resolved": { "name", "primary_role", "primary_company", "confidence" },
  "profiles": [{ "platform", "handle", "url", "verified" }],
  "latest_activity": { "platform", "date", "snippet", "link" },
  "candidates": ["…"],
  "credits_used": 2
}

Example invocations

1. Claude Desktop

Find Maya Patel at Linear. I need her X handle and most
recent post.

2. ChatGPT custom GPT

Before writing any outbound, call find_contact to confirm
the right profiles. Never guess handles.

3. Cursor MCP

Resolve these 30 names+companies with find_contact.
Output a CSV of name, company, x_handle, linkedin_url.

4. n8n

{ "jsonrpc": "2.0", "method": "tools/call",
  "params": { "name": "find_contact",
    "arguments": { "name": "Lenny Rachitsky" } } }

5. Raw curl

curl -X POST https://www.mentionfox.com/mcp \
  -H "Authorization: Bearer $FOXAPIS_KEY" \
  -d '{"jsonrpc":"2.0","method":"tools/call","params":{"name":"find_contact","arguments":{"name":"Guillermo Rauch","company":"Vercel"}}}'

Sample output

{
  "resolved": { "name": "Guillermo Rauch", "primary_role": "CEO", "primary_company": "Vercel", "confidence": 0.99 },
  "profiles": [
    { "platform": "x", "handle": "@rauchg", "verified": true },
    { "platform": "github", "handle": "rauchg" }
  ],
  "credits_used": 2
}

Credit cost & rate limits

Flat 2 credits. 90 calls per minute per key. Bulk lookups should be done as a stream, not a parallel burst.

Error codes & recovery