sec-finance-ai MCP server

AI-powered SEC EDGAR integration: 23 tools on the official sec.gov APIs, in one Python file.

🚀 Thousands of downloads on ClawHub and OpenWebUI

10-K, 10-Q, 8-K, 20-F and proxy filings, the full text of any filing read in pages or searched, XBRL company facts and cross-company rankings, Form 4 insider trades, 13D/13G ownership, 13F fund portfolios, IPOs, and full-text search of every filing since 2001.

Install

Needs uv. No API key.

Claude Code:

claude mcp add sec-ai -e SEC_USER_AGENT="Your Name you@example.com" -- uvx sec-finance-ai@latest

VS Code:

code --add-mcp '{"name":"sec-ai","type":"stdio","command":"uvx","args":["sec-finance-ai@latest"],"env":{"SEC_USER_AGENT":"Your Name you@example.com"}}'

Claude Desktop, Cursor, Windsurf and other clients, in the client's MCP config file:

{
  "mcpServers": {
    "sec-ai": {
      "command": "uvx",
      "args": [
        "sec-finance-ai@latest"
      ],
      "env": {
        "SEC_USER_AGENT": "Your Name you@example.com"
      }
    }
  }
}

@latest picks up new releases when the client starts. Without uv: pip install sec-finance-ai, then use "command": "sec-finance-ai" with no args.

SEC_USER_AGENT is the name and e-mail the SEC asks programs to send. If it is not set, a default is sent.

Listed in the MCP Registry as io.github.lkcair/sec-ai. Every tool is read-only.

Agent frameworks

The server runs over stdio with uvx sec-finance-ai@latest, so any framework with MCP support can load all its tools.

SEC_USER_AGENT is optional. MCP starts the server with a minimal environment, so pass it as shown; **os.environ keeps PATH, so uvx is found.

LangChain: pip install "langchain[mcp]" langchain-anthropic

import asyncio
import os
from langchain.agents import create_agent
from langchain.mcp import MCPAdapter

async def main():
    config = {"mcpServers": {"sec-ai": {
        "command": "uvx", "args": ["sec-finance-ai@latest"],
        "env": {**os.environ, "SEC_USER_AGENT": "Your Name you@example.com"},  # optional
    }}}
    async with MCPAdapter(config) as adapter:
        tools = await adapter.list_tools()
        agent = create_agent("claude-sonnet-5", tools)
        result = await agent.ainvoke({"messages": [{"role": "user", "content": "What does Apple's latest 10-K say about Greater China?"}]})
        print(result["messages"][-1].content)

asyncio.run(main())

Google ADK: pip install "google-adk[mcp]"

import os
from google.adk.agents import LlmAgent
from google.adk.tools.mcp_tool import McpToolset
from google.adk.tools.mcp_tool.mcp_session_manager import StdioConnectionParams
from mcp import StdioServerParameters

root_agent = LlmAgent(
    model="gemini-flash-latest",
    name="sec_ai",
    instruction="Answer with the sec-ai tools.",
    tools=[McpToolset(connection_params=StdioConnectionParams(
        server_params=StdioServerParameters(
            command="uvx", args=["sec-finance-ai@latest"],
            env={**os.environ, "SEC_USER_AGENT": "Your Name you@example.com"},  # optional
        ),
        timeout=60,
    ))],
)

Strands Agents: pip install strands-agents

import os
from mcp import stdio_client, StdioServerParameters
from strands import Agent
from strands.tools.mcp import MCPClient

sec_ai = MCPClient(lambda: stdio_client(StdioServerParameters(
    command="uvx", args=["sec-finance-ai@latest"],
    env={**os.environ, "SEC_USER_AGENT": "Your Name you@example.com"},  # optional
)))
agent = Agent(tools=[sec_ai])
agent("What does Apple's latest 10-K say about Greater China?")

Example questions

• What does Apple's latest 10-K say about Greater China?
• What are Berkshire Hathaway's top holdings in its latest 13F?
• Find every 8-K this year that mentions 'going concern'.
• Which companies had the highest revenue in 2024?

MCP tools (22)

Filings

Reading inside a filing

Search

Ownership and insiders

Financial data (XBRL)

Utility

License

MIT. All data comes from public SEC EDGAR databases. Not affiliated with the U.S. Securities and Exchange Commission.