XFINLAB

Financial Intelligence Infrastructure — APIs, SDKs, and an MCP server for developers and AI agents, plus a consumer research platform built on the same backend.

Real market events, FinBERT sentiment, technical/market-structure analysis, SEC/CFTC/FDIC/USDA/CBOE official data, and Monte Carlo stress testing. Every field is traceable to a real computation or an official data source — nothing fabricated or interpolated.

Get a free API key · API docs · llms.txt · Consumer product

Quick start (API)

pip install "git+https://github.com/lnanology/Xfinlab.git#subdirectory=sdk/python"

from xfinlab_intelligence import XfinlabClient

client = XfinlabClient(api_key="xfl_...")  # free tier, issued instantly
sentiment = client.sentiment("AAPL")
technical = client.technical("AAPL", period="6mo")
fundamentals = client.fundamentals("AAPL")

Or JavaScript/Node:

npm install "github:lnanology/Xfinlab#path:sdk/js"

const { XfinlabClient } = require('xfinlab-intelligence');
const client = new XfinlabClient('xfl_...');
const sentiment = await client.sentiment('AAPL');

19 endpoints total — market events, sentiment, AI debate, technical/market-structure, Monte Carlo stress testing, insider trading, institutional ownership, short interest, SEC XBRL fundamentals, CBOE VIX term structure, FDIC bank health, USDA agriculture, EIA energy, crypto, cross-region market map, and Pro-tier webhooks. Full reference: intelligence-api.html.

MCP server (for Claude and other AI agents)

Every field this server returns is either a real computation/real official-source value, or null with an explanation — the MCP ecosystem has a lot of servers now, few of them say anything about the actual quality of the data behind the tool calls. This one does, in public: xfinlab.com/trust.html shows every underlying data collector's live/down status in real time.

Already live in production — no setup needed, just point an MCP-compatible client at it:

{
  "mcpServers": {
    "xfinlab": {
      "url": "https://api.xfinlab.com/api/mcp",
      "headers": { "X-API-Key": "xfl_..." }
    }
  }
}

Server source: api/mcp_server.py. Tools: get_market_events, get_sentiment, get_technical_analysis, get_intelligence_feed, get_global_market_map. Same auth and free tier as the REST API. Docs: intelligence-api.html#mcp.

SDKs & examples

Both SDKs are MIT-licensed (sdk/LICENSE) and have zero required dependencies beyond the standard library / native fetch.

Consumer product

The same backend also powers xfinlab.com, a retail investment-research platform:

Local development

python3 mock-server.py

Then open http://localhost:8080. The production backend is a separate FastAPI app (backend/main.py, deployed on Railway as api.xfinlab.com); the static site above deploys separately on Vercel as xfinlab.com.

More docs

• XFINLAB_ARCHITECTURE.md — full system architecture
• PROJECT_ROADMAP.md — roadmap
• PROJECT_STYLE_GUIDE.md — UI style guide
• DATA-LICENSE-MATRIX.md — upstream data source licensing for every paid endpoint
• MCP_MARKETPLACE_SUBMISSION.md — MCP server directory submission copy