Best for
- Researchers who need AI-assisted access to U.S. public-company filings
- Users analyzing company, financial, or insider-trading data
- Financial research workflows that require precise numbers and source links
S
SEC EDGAR MCP Server
Community
Give AI assistants access to SEC EDGAR filings.
- Category
- Other #83 of 230
- Stars
- ★ 373 Very popular
- Transport
- stdio (local process) · Streamable HTTP
- Runtime
- Docker · Python 3.11+
- Credentials
- No credential needed
- License
- AGPL-3.0
- Last commit
59FMRS · C
This active AGPL-3.0 server provides access to company filings, financial data, and insider transactions through stdio and streamable HTTP. The source does not specify exact tool identifiers, so none are inferred.
Strongest · Setup experience 13/20 Weakest · Reliability 10/20
Why each score
- Reliability 10/20
- The manifest declares stdio transport, a required SEC_EDGAR_USER_AGENT, and company, filing, financial, and insider-trading tools. The README also mentions Promptfoo evaluations and a CI workflow, making the normal path plausible. However, only the manifest and README are available here: there is no source code, test content, error-handling evidence, or observed handshake. Tool behavior, dependency control, and edge cases therefore remain unverified; the score is capped conservatively below 12 for static review.
- Security and permissions 11/20
- The evidence indicates primarily read-only access to public SEC EDGAR data, with a non-secret user-agent variable and no declared write or destructive operations. The README explicitly discloses that streamable HTTP has no authentication and should be used only on private networks. However, external network behavior, data flows, rate limits, input scoping, and deployment isolation are not documented; unauthenticated HTTP leaves access control incomplete. No stated red-line risk is evident.
- Maintenance 13/20
- The repository is not archived and provides a 1.0.5 manifest version, PyPI and Conda distribution paths, a GitHub repository, an AGPL-3.0 license, maintainer contact information, a CI evaluation workflow, and basic traceability. The README cites version 1.0.6 while the manifest says 1.0.5, creating a release inconsistency. Commit history, dependency-update policy, vulnerability response process, and issue-response evidence are not supplied, so substantial points are deducted.
- Documentation 12/20
- The README explains the purpose, main tool categories, Docker quick start, stdio requirement, HTTP launch mode, lack of authentication, installation channels, evaluation entry point, documentation link, SEC non-endorsement, and license. It does not provide reviewable tool parameters, response schemas, error behavior, SEC rate limits, cost or network-dependency details, compatibility guidance, or troubleshooting. Important details are deferred to external documentation, so the score is moderate rather than high.
- Setup experience 13/20
- A usable Docker client configuration is provided, including the required -i flag, and pip, Conda, uv, and streamable HTTP paths are identified. Python 3.11+ is also signaled, so the basic stdio setup appears relatively short. Still, there is no complete local-install configuration, client compatibility matrix, connection-validation step, prerequisite checklist, or common-failure guidance; HTTP deployment also requires the user to secure the private network. These omissions reduce the score.
Static review · not runListed 2026-08-14
Read the FMRS scoring method →Fit and risk
What it can accessUses the network
Not for
- Users requiring an SEC-maintained or SEC-endorsed server
- Public HTTP deployments requiring built-in authentication
- Other data sources or financial-service capabilities not described in the source material
Required permissions
- Requires access to the SEC EDGAR API
- Requires the non-secret SEC_EDGAR_USER_AGENT environment variable
- HTTP mode should run on a controlled network because authentication is not included
Risks and side effects
- The project states that it is not affiliated with or endorsed by the U.S. Securities and Exchange Commission
- HTTP transport has no authentication and should not be exposed to untrusted networks
- SEC API access may depend on correct user-agent configuration and external service availability
- Financial and insider-trading information should be verified through the returned SEC filing URLs
Setup
Before you start
Runtime：Docker · Python 3.11+
SEC_EDGAR_USER_AGENT
Required user agent for the SEC EDGAR API, formatted as 'Your Name ([email protected])' — use your own name and email; no registration needed.
Run it over stdio with Docker. The configuration must set SEC_EDGAR_USER_AGENT to a user-agent string and use docker run -i --rm. Pip, conda, and uv installation methods are also supported according to the project documentation. For platforms such as Dify, use streamable HTTP with: python -m sec_edgar_mcp.server --transport streamable-http --port 9870.
MCP config · JSON
{"mcpServers":{"sec-edgar-mcp":{"command":"docker","args":["run","-i","--rm","-e","SEC_EDGAR_USER_AGENT=Your Name ([email protected])","stefanoamorelli/sec-edgar-mcp:latest"]}}}
Works in clients that read an "mcpServers" key. For clients such as VS Code that use a different key, the configurator below converts it.
.vscode/mcp.json
{
"servers": {
"sec-edgar-mcp": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"SEC_EDGAR_USER_AGENT=Your Name ([email protected])",
"stefanoamorelli/sec-edgar-mcp:latest"
]
}
}
}
Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).
Terminal
claude mcp add sec-edgar-mcp -- docker run -i --rm -e 'SEC_EDGAR_USER_AGENT=Your Name ([email protected])' stefanoamorelli/sec-edgar-mcp:latest
Run it in a terminal; replace any <…> placeholders with your own values first.
Check that it works
After launching sec-edgar-mcp in your client, company info and filing tools should appear in the tool list; asking "Look up Apple's CIK number" should return SEC EDGAR data with filing URLs, proving the connection works.
Troubleshooting
- Confirm that SEC_EDGAR_USER_AGENT is set with identifiable user-agent information
- Ensure the Docker stdio configuration includes -i for MCP JSON-RPC communication
- For HTTP mode, confirm that port 9870 is reachable and the deployment uses a private network
- Check network connectivity to the SEC EDGAR API
- Verify that the selected pip, conda, or uv installation method matches the project documentation
Things to try
Once connected, you can ask your AI assistant things like:
- Look up Apple's CIK number and company info on SEC EDGAR
- Fetch Apple's latest 10-K filing and extract the business description section
- Show Tesla's balance sheet from the most recent quarter
- List recent Form 4 insider trading transactions for Nvidia
Use cases
Look up CIKs, company information, and company facts
Retrieve 10-K, 10-Q, and 8-K filings and extract sections
Access XBRL-parsed balance sheets, income statements, and cash-flow statements
Query Form 3, 4, and 5 insider transactions
Verify results using the included SEC filing URLs
Supported clients
DifyPartial support
Listed from the project's documentation, not tested by this site.
Overview
SEC EDGAR MCP Server gives AI assistants access to public filings through the U.S. Securities and Exchange Commission EDGAR API. It covers company filings, financial statements, and insider trading data, with an emphasis on exact numeric precision. Built on edgartools, it includes SEC filing URLs for verification.
Similar servers
Hevy MCP
77 · B
Manage your Hevy workout data from AI assistants via MCP.
Firecrawl MCP Server
75 · B
Firecrawl's official MCP server for web search, scraping, and structured extraction for AI agents
BioMCP
75 · B
One binary. One grammar. Evidence from the biomedical sources you already trust.
Compartment
75 · B
Encrypted, fully offline long-term memory for AI agents
Source revision 86646eb0e23f Data synced 2026-10-19 Read the FMRS scoring method