get_dossierGenerate a full dossier on any subject.

A vet is a paragraph. A dossier is a document. get_dossier takes a subject and a depth setting and assembles a multi-source profile — bio, employment, recent activity, mentions, comp signal, sentiment, network, risks — into a single structured object the agent can hand to a human or paste into a CRM. Fifteen credits. About fifteen seconds. Designed for the moments when "find out everything we can" is the actual brief.

When to call this tool

An agent should reach for get_dossier when the downstream consumer is a human reading a document, not the agent itself reasoning over a small object. Investor diligence calls. Pre-meeting briefs for an exec. A journalist's source-check. A recruiter's full read on a finalist before reference calls. The tool returns a structure that maps cleanly onto a one-page brief, including a synthesised narrative section that reads like an analyst wrote it. Every claim cites its source.

Compared to vet_person, the dossier is roughly three times the depth and triggers an internal mention-scan, a comp-signal pull, and a sentiment pass over the last 90 days of activity. It is the right tool whenever you would otherwise tell a junior person to "spend an hour on this".

Input schema

{
  "subject": "string  // name, handle, or domain",
  "subject_kind": "enum: 'person' | 'company' | 'auto'",
  "depth": "enum: 'standard' | 'deep' | 'forensic'  // default 'standard'",
  "sections": ["string  // optional, restrict to subset"],
  "include_network": "boolean  // default false; adds 5 cr"
}

Output schema

{
  "subject": { "name", "resolved", "confidence" },
  "narrative": "string  // 6-12 paragraph synthesis",
  "bio": { "role", "company", "location", "tenure" },
  "employment_history": [{ "company", "role", "start", "end" }],
  "recent_activity": [{ "date", "platform", "summary", "link" }],
  "mentions_summary": { "count_90d", "top_platforms", "sentiment" },
  "comp_signal": { "band", "basis" },
  "network": [{ "name", "relation", "strength" }],
  "risks": [{ "type", "severity", "detail", "source" }],
  "sources": ["url"],
  "credits_used": 15
}

Example invocations

1. Claude Desktop

// User prompt
Build me a full dossier on the founder of Stripe before
my 9am with him tomorrow. Deep depth, include network.

2. ChatGPT custom GPT

// GPT instructions
For any "brief me on X" or "tell me everything about X"
prompt, call get_dossier with depth: 'deep'. Open your
reply with the narrative section.

3. Cursor MCP

// In Cursor
Use get_dossier to write a markdown one-pager for
"Patrick Collison". Save to ./briefs/collison.md.

4. n8n

{ "jsonrpc": "2.0", "id": 1, "method": "tools/call",
  "params": { "name": "get_dossier",
    "arguments": { "subject": "Vercel", "subject_kind": "company", "depth": "deep" } } }

5. Raw curl

curl -X POST https://www.mentionfox.com/mcp \
  -H "Authorization: Bearer $FOXAPIS_KEY" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"get_dossier","arguments":{"subject":"Naval Ravikant","depth":"standard"}}}'

Sample output (real, redacted)

{
  "subject": { "name": "Naval Ravikant", "confidence": 0.99 },
  "narrative": "Founder of AngelList. Investor in 200+ companies including Twitter, Uber, Stripe...",
  "bio": { "role": "Founder, AngelList", "location": "San Francisco" },
  "recent_activity": [{ "date": "2026-04-22", "platform": "x", "summary": "Thread on AI as a creative partner..." }],
  "mentions_summary": { "count_90d": 3417, "sentiment": "positive" },
  "credits_used": 15
}

Credit cost & rate limits

Standard depth costs 15 credits. Deep is 25. Forensic is 40. Adding include_network: true adds 5. Rate limit is 15 per minute, with a daily soft cap of 200 dossiers per key.

Error codes & recovery