federal-spending-mcp

MCP server that gives AI assistants direct access to federal contract intelligence from USASpending.gov. No API key needed — all public data.

18 tools covering every federal agency: search awards, analyze spending, profile contractors, and run full BD workflows that produce actionable reports.

Install

Claude Desktop

Add to ~/.claude/claude_desktop_config.json:

{
  "mcpServers": {
    "usaspending": {
      "command": "npx",
      "args": ["-y", "federal-spending-mcp"]
    }
  }
}

Claude Code

Add to your project's .mcp.json:

{
  "mcpServers": {
    "usaspending": {
      "command": "npx",
      "args": ["-y", "federal-spending-mcp"]
    }
  }
}

Or install globally:

claude mcp add usaspending -- npx -y federal-spending-mcp

Remote (hosted — no install needed)

If you're running the hosted version, just paste a URL:

{
  "mcpServers": {
    "federal-spending": {
      "url": "https://federal-spending-mcp.onrender.com/sse"
    }
  }
}

Cursor / Windsurf / Other MCP Clients

Same pattern — point your MCP config at npx -y federal-spending-mcp or use the hosted URL above.

Self-Hosting

Deploy your own instance with Docker:

docker build -t federal-spending-mcp .
docker run -p 3000:3000 federal-spending-mcp

Or deploy to Render:

1. Fork this repo
2. Create a new Web Service on Render
3. Point it to your fork
4. Set environment to Docker
5. It auto-deploys — SSE endpoint is at https://your-service.onrender.com/sse

Environment variables:

• PORT — HTTP port (default: 3000, Render sets this automatically)
• MCP_TRANSPORT — http for remote, stdio for local (auto-detected)

Tools

Data Tools (12)

| Tool | Description | |---|---| | list_federal_agencies | All federal agencies with budget authority and obligations | | search_awards | Search contracts by keyword, agency, location, NAICS/PSC, recipient, date | | get_agency_spending | Spending breakdown by agency or sub-agency | | get_top_recipients | Top contractors ranked by award value with HHI concentration | | get_sector_spending | NAICS industry and PSC product/service breakdowns | | get_geographic_spending | Spending by state/territory | | get_spending_trends | Monthly/quarterly/yearly trends with YoY growth | | get_agency_overview | Detailed agency profile by toptier code | | analyze_opportunity_landscape | Hot sectors, active agencies, market concentration | | profile_location | Full contracting profile for a city/state | | profile_recipient | Deep dive on a specific contractor | | track_spending_filtered | Spending over time with location/keyword/agency filters |

Workflow Tools (6)

These solve the cold-start problem — minimal input, full report output.

| Tool | Input | Output | |---|---|---| | opportunity_scan | Your capabilities (keywords, NAICS) | Where matching federal dollars flow, incumbents, entry difficulty, trend analysis | | competitive_intel | Competitor company name | Federal footprint, agency diversification, concentration risk, growth trajectory | | agency_penetration_map | Target agency | Budget breakdown, vendor lock-in, niche opportunities, emerging signals | | territory_briefing | State/city | Market size, local competitors, spending momentum, sector gaps | | pipeline_forecast | Agency or sector | YoY trends, acceleration detection, quarterly momentum, forecast signals | | capture_readiness | Your company + target agency | Readiness score (0-100), capability gaps, teaming candidates, action plan |

Example Queries

Once installed, just ask your AI assistant:

• "What are the top DoD contractors?"
• "Search for cybersecurity contracts at DHS"
• "Run a competitive intel report on Booz Allen Hamilton"
• "Show me NASA's spending trends over the last 5 years"
• "Give me a territory briefing for San Antonio, TX"
• "Assess my company's capture readiness for Department of Energy contracts"
• "What's the pipeline forecast for Department of Veterans Affairs?"
• "Find contracts related to artificial intelligence at all agencies"
• "Profile the Warner Robins, GA contracting landscape"
• "Who are the top recipients of Department of Energy funding?"

Example Output

Competitive Intelligence: Booz Allen Hamilton

Total Federal Value: $23.7B
Contracts Analyzed: 50
Agencies Served: 11
Market Rank: #10

Concentration Risk: MODERATE
  Top Agency: Federal Acquisition Service (53.2%)

Capabilities: Data processing, IT services, management consulting
Growth: Tracked across 11 sub-agencies

Capture Readiness: Palantir → Department of Veterans Affairs

Readiness Score: 90/100
Level: HIGH — Strong position to compete. Proceed with capture.
Existing Presence: Yes
Federal Experience: $2.3B
Matching NAICS: 2

Recommended Actions:
  → Identify specific recompetes from the awards list
  → Build relationships with contracting officers
  → Develop past performance narratives aligned to agency mission

Pipeline Forecast: NASA

Direction: GROWING
Accelerating: Yes
Quarterly Momentum: +53.4%

YoY Growth:
  FY2021→FY2022: +1.4%
  FY2022→FY2023: +6.0%
  FY2023→FY2024: -6.9%
  FY2024→FY2025: +5.1%

Signal: Budget acceleration detected — spending is growing
and the rate of growth is increasing.

How It Works

All data comes from the USASpending.gov API, which is public and requires no authentication. The server makes real-time queries — no data is cached or stored.

Workflow tools make 5-10 parallel API calls each and synthesize the results into structured reports.

For OpenAI / Other LLMs

The openai-tools.json file in the repo contains all tool definitions in OpenAI function-calling format. Use it as the tools parameter in chat completions.

License

MIT