GrantIQ MCP
A paid MCP server that gives AI agents instant, structured access to every open US federal grant and contract opportunity.
README
GrantIQ MCP
A paid MCP server that gives AI agents instant, structured access to every open US federal grant and contract opportunity. Backed by Grants.gov + SAM.gov, with 24-hour SQLite caching, per-key daily limits, and a Stripe-friendly upgrade hook.
https://mcp-grantiq.up.railway.app/mcp
MCPize listing copy
Title: GrantIQ MCP — US Federal Grants & Contracts for AI Agents
Subtitle: Search, match, and monitor every open SAM.gov and Grants.gov opportunity from any LLM, in one tool call.
GrantIQ MCP turns the messy world of US government funding into a clean, agent-ready API. Eight typed tools cover everything from broad keyword search and NAICS-coded contract discovery to deadline radar and historical award lookups. Responses are normalized across both data sources, returned as plain JSON, and cached for 24 hours so repeat queries don't burn rate limit on the upstream agencies.
Built for procurement copilots, grant-finder agents, and B2G sales tools, GrantIQ MCP ships with a free tier (50 calls/day) that any developer can drop into Claude or Cursor in 15 seconds. Upgrade to Pro for $19/mo for unlimited calls, or pay $0.02 per call for bursty workloads — billing flips a key from free to pro via a single webhook, with no app rewrites required.
Pricing
| Tier | Price | Limit |
|---|---|---|
| Free | $0 | 50 calls / day / key |
| Pro | $19 / month | Unlimited |
| Pay-as-you-go | $0.02 / call | Unlimited |
Install
claude mcp add grantiq-mcp --url https://mcp-grantiq.up.railway.app/mcp
Then add the API key Claude should use:
claude mcp env grantiq-mcp X-API-Key=grantiq-dev-key-001
The dev key grantiq-dev-key-001 is seeded automatically and capped at 50 calls/day. Mint your own with the admin endpoint below.
Tool reference
| Tool | Args | Returns |
|---|---|---|
search_grants |
keyword, agency, amount_min, amount_max, status, limit |
List of matching grants |
get_grant_details |
opportunity_id |
Full grant object |
search_contracts |
keyword, naics_code, agency, set_aside, limit |
List of SAM contracts |
get_contract_details |
notice_id |
Full contract object |
match_opportunities |
org_description, org_type, focus_areas |
Top 10 matched opps (grants + contracts) |
get_agencies |
type (grants / contracts / both) |
Agencies with open opps, ranked by volume |
get_deadlines |
days_ahead |
Opportunities closing within N days |
search_awards |
recipient_name, agency, year |
Past SAM award history |
Every tool returns either:
{ "success": true, "cached": false, "data": { ... } }
or, on failure:
{ "success": false, "error": "..." }
When the daily limit is hit, the HTTP layer returns 429 with body:
{ "error": "Limit reached. Upgrade at mcpize.com/grantiq-mcp" }
curl examples
The MCP endpoint speaks the JSON-RPC streamable-HTTP protocol and is best used through an MCP client. For convenience, every tool also has a REST mirror at /tools/<tool>.
# Search grants
curl -s https://mcp-grantiq.up.railway.app/tools/search_grants \
-H 'Content-Type: application/json' \
-H 'X-API-Key: grantiq-dev-key-001' \
-d '{"keyword":"renewable energy","limit":5}'
# Get grant details
curl -s https://mcp-grantiq.up.railway.app/tools/get_grant_details \
-H 'Content-Type: application/json' \
-H 'X-API-Key: grantiq-dev-key-001' \
-d '{"opportunity_id":"345678"}'
# Search SAM.gov contracts (cyber, small-business set-aside)
curl -s https://mcp-grantiq.up.railway.app/tools/search_contracts \
-H 'Content-Type: application/json' \
-H 'X-API-Key: grantiq-dev-key-001' \
-d '{"keyword":"cybersecurity","set_aside":"SBA","limit":5}'
# Get contract details
curl -s https://mcp-grantiq.up.railway.app/tools/get_contract_details \
-H 'Content-Type: application/json' \
-H 'X-API-Key: grantiq-dev-key-001' \
-d '{"notice_id":"abc123"}'
# Match opportunities to an org profile
curl -s https://mcp-grantiq.up.railway.app/tools/match_opportunities \
-H 'Content-Type: application/json' \
-H 'X-API-Key: grantiq-dev-key-001' \
-d '{"org_description":"Battery storage R&D startup","org_type":"for-profit","focus_areas":["energy","battery","storage"]}'
# Top agencies right now
curl -s https://mcp-grantiq.up.railway.app/tools/get_agencies \
-H 'Content-Type: application/json' \
-H 'X-API-Key: grantiq-dev-key-001' \
-d '{"type":"both"}'
# Deadlines in the next 14 days
curl -s https://mcp-grantiq.up.railway.app/tools/get_deadlines \
-H 'Content-Type: application/json' \
-H 'X-API-Key: grantiq-dev-key-001' \
-d '{"days_ahead":14}'
# Past award history
curl -s https://mcp-grantiq.up.railway.app/tools/search_awards \
-H 'Content-Type: application/json' \
-H 'X-API-Key: grantiq-dev-key-001' \
-d '{"recipient_name":"Acme Corp","year":2025}'
Operator endpoints
Mint a fresh key (free or pro):
curl -s https://mcp-grantiq.up.railway.app/admin/keys \
-H 'X-Admin-Token: $GRANTIQ_ADMIN_TOKEN' \
-H 'Content-Type: application/json' \
-d '{"tier":"free","daily_limit":50}'
Promote a free key to pro (called from the Stripe checkout success webhook):
curl -s https://mcp-grantiq.up.railway.app/billing/upgrade \
-H 'X-Admin-Token: $GRANTIQ_ADMIN_TOKEN' \
-H 'Content-Type: application/json' \
-d '{"key":"grantiq_xxxxx"}'
Local dev
python -m venv .venv && source .venv/bin/activate
pip install -r requirements.txt
cp .env.example .env
python server.py # http://localhost:8000
pytest -v # 22 tests, hermetic with respx mocks
Deploy to Railway
The repo includes nixpacks.toml. After railway init, just railway up. The start command is python server.py (not uvicorn ...) because Railway exec's [start].cmd without a shell — server.py reads $PORT via os.getenv and starts uvicorn programmatically.
Set these on the Railway service:
SAM_API_KEY— your SAM.gov keyGRANTIQ_ADMIN_TOKEN— long random stringGRANTIQ_DB_PATH—/data/grantiq.dbif you mount a Volume (recommended so cache + keys survive deploys)
A ready-to-run deploy.sh is in this directory.
Caveats
- SAM.gov award history is exposed via the opportunities API filtered to
ptype=a(Award Notice). For deeper award analytics (CFDA-level totals, recipient rollups), USAspending.gov is more appropriate and would be a v2 add. - Cache TTL is global (24h). For deadline-sensitive flows the agent should pass distinct query args or call
/admin/keysto mint a key that bypasses cache (not yet exposed — see TODO). - Match scoring is deterministic Jaccard over title + agency tokens. It's fast and free; an LLM rerank can be added behind the same interface.
- Stripe wiring is one-way (
/billing/upgrade). The MCP server doesn't host the checkout page itself — that lives on the marketing site.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。