Search FDIC-insured banks, financials, failures, and risk signals via fdic-mcp-server (npx)

Recipe: search FDIC-insured banking institutions via fdic-mcp-server

Server: [email protected] (npm, stdio) Launch: npx -y [email protected] Auth: none — FDIC BankFind Suite API is public US government data Tools: 29+ including search, financials, failures, CAMELS health, risk signals, peer analysis

What it does

Wraps the entire FDIC BankFind Suite API as an MCP server. Query FDIC-insured banking institutions by name, state, certificate number, or asset size. Get quarterly financials, track bank failures, run peer-group comparisons, and detect early-warning risk signals — all from public government data.

MCP handshake

→ initialize(protocolVersion="2024-11-05")
← serverInfo: {name:"fdic-mcp-server", version:"1.30.0"}
← capabilities: {tools:{listChanged:true}, resources:{listChanged:true}, prompts:{listChanged:true}}

→ tools/list
← 29+ tools including:
  - fdic_search_institutions: search by NAME, STALP, CERT, ASSET, ACTIVE
  - fdic_get_institution: get one institution by CERT number
  - fdic_search_failures: search failed institutions
  - fdic_search_financials: quarterly Call Report data
  - fdic_analyze_bank_health: CAMELS-style assessment
  - fdic_detect_risk_signals: early-warning scanning
  - fdic_peer_group_analysis: rank against peers
  - fdic_market_share_analysis: deposit market share by MSA

Real execution: largest banks in New York state

→ tools/call  name="fdic_search_institutions"
  arguments={
    "filters": "STALP:NY AND ACTIVE:1",
    "fields": "NAME,CERT,ASSET,DEP,STALP,CITY",
    "limit": 5,
    "sort_by": "ASSET",
    "sort_order": "DESC"
  }
← {
    total: 119,
    count: 5,
    has_more: true,
    next_offset: 5,
    institutions: [
      {NAME:"Goldman Sachs Bank USA", CERT:33124, ASSET:751776000, DEP:504823000, CITY:"New York"},
      {NAME:"The Bank of New York Mellon", CERT:639, ASSET:467349000, DEP:419724000, CITY:"New York"},
      {NAME:"Morgan Stanley Private Bank, National Association", CERT:34221, ASSET:241388000, DEP:205802000, CITY:"Purchase"},
      {NAME:"Manufacturers and Traders Trust Company", CERT:588, ASSET:214201000, DEP:167256000, CITY:"Buffalo"},
      {NAME:"Flagstar Bank, National Association", CERT:32541, ASSET:87128815, DEP:66832203, CITY:"Hicksville"}
    ]
  }

Response time: ~1-2 seconds (network call to FDIC API).

Filter syntax

The filters parameter uses ElasticSearch query-string syntax:

• NAME:Chase — name contains "Chase"
• STALP:CA AND ACTIVE:1 — active California banks
• ASSET:[1000000 TO *] — assets > $1B (ASSET is in $thousands)
• Combine with AND / OR, use quotes for multi-word values

ASSET units

ASSET and DEP values are in thousands of dollars. Goldman Sachs at 751,776,000 = $751.8 billion.

Gotchas

• Name searches are case-sensitive and exact within quotes — use unquoted partial matches or state filters instead
• The server also exposes MCP resources and prompts (not just tools)
• Advanced tools like fdic_analyze_bank_health and fdic_detect_risk_signals combine multiple FDIC endpoints into synthesized analysis