sec-filings-rag-agent

animesh01/sec-filings-rag-agent
★ 0 stars Python AI/LLM Updated 2d ago
Finance-ops RAG agent over two sources — real SEC EDGAR filings and FP&A planning documents — with answers grounded in retrieved passages, citations to the exact section, and a refusal when out-of-corpus. Demonstrates an end-to-end RAG pipeline, token minimization, and MCP-server retrieval tools. Built on public SEC data plus synthetic FP&A docs.
View on GitHub → Try with Claude — $10 free →

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "sec-filings-rag-agen": {
      "command": "uvx",
      "args": [
        "sec-filings-rag-agent"
      ]
    }
  }
}

Or install with pip: pip install sec-filings-rag-agent

README Excerpt

> Ask a finance question, get an answer **grounded in retrieved document passages** — across **two sources**: > real **SEC EDGAR filings** and **FP&A planning documents** (variance commentary, forecast assumptions, close > notes, policy). Every answer cites the exact document and section, and the agent **refuses** rather than guess

Topics

agentic-aiai-agentscitationscontext-protocolfinancefpagroundingnlpretrieval-augmented-generationsec-edgarstreamlittoken-optimizationvector-search