Equibles

Name: Equibles
Author: daniel3303

@daniel3303

Equibles について

Self-hosted financial data terminal for AI agents. Scrapes and serves SEC filings (full-text search), 13F institutional holdings, insider and congressional trades, FINRA short data, FRED economic indicators, CFTC futures positioning, CBOE VIX/put-call ratios, and daily stock pric

#finance #stocks #sec-filings #insider-trading #self-hosted

基本情報

カテゴリ

金融とコマース

トランスポート

stdio

公開者

daniel3303

投稿者

Daniel Oliveira

設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "equibles": {
      "transport": "streamable-http",
      "url": "https://mcp.equibles.com/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    },
    "equibles-local": {
      "url": "http://localhost:8081/mcp"
    }
  }
}

ツール

ツールは検出されませんでした

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is Equibles?

Equibles is an open-source, self-hosted mini Bloomberg Terminal for AI agents. It scrapes, stores, and serves SEC filings, institutional holdings, insider trading, congressional trades, short data, economic indicators, and daily stock prices — and exposes everything via the Model Context Protocol (MCP) so AI assistants like Claude and ChatGPT can query it directly.

How to use Equibles?

Run git clone https://github.com/daniel3303/Equibles.git, copy .env.example to .env, set SEC_CONTACT_EMAIL, then docker compose up. The MCP server starts on port 8081. Connect any MCP‑compatible client (e.g., Claude Desktop, ChatGPT Desktop, or Claude Code) to http://localhost:8081/mcp. Optional free API keys for FINRA short data and FRED economic data can be set in .env.

Key features of Equibles

Scrapes and stores SEC filings (10‑K, 10‑Q, 8‑K) with full‑text search
Tracks institutional holdings (13F‑HR) and insider trading (Form 3/4)
Monitors congressional stock trades from House/Senate disclosures
Serves short data from SEC fails‑to‑deliver and FINRA short volume/interest
Provides economic indicators from FRED and daily stock prices from Yahoo Finance
Exposes all data via an MCP server for AI assistants

Use cases of Equibles

Ask your AI assistant “Who are the top institutional holders of AAPL?”
Search the latest 10‑K for revenue growth discussion
Retrieve insider trading transactions for a specific company
Get a macro snapshot with the latest FRED economic indicators
Check VIX volatility history and put/call ratios by category

FAQ from Equibles

What are the runtime dependencies?

Docker and Docker Compose are required for the recommended setup. The application targets .NET 10. ParadeDB (PostgreSQL + pgvector + pg_search) is used as the database.

Where does the data come from and live?

Data is scraped from public sources: SEC EDGAR, FINRA, FRED, Yahoo Finance, CFTC, and CBOE. All data is stored locally in a ParadeDB instance (port 5432) inside the Docker stack.

Does this require any API keys?

A free FINRA API key (Client ID/Secret) is needed for short volume and short interest data. A free FRED API key is required for economic indicators. Both are optional — scrapers skip gracefully if keys are missing. An SEC EDGAR contact email is required for fair access policy.

How do I connect Equibles to an AI assistant?

For Claude Desktop, add an entry with "url": "http://localhost:8081/mcp" to the mcpServers object in the config file. For ChatGPT Desktop, add to servers in mcp.json. Other MCP clients use the same HTTP endpoint.

What transport and authentication are supported?

The MCP server uses HTTP transport. Authentication is optional: set MCP_API_KEY in .env to require a Bearer token on requests. The web portal can be password‑protected with AUTH_USERNAME and AUTH_PASSWORD.