Insider Trading MCP Server

@pipeworx/insider-trading

SEC EDGAR insider and ownership filings for US-listed companies — Form 4 insider transactions, 8-K material events, and Schedule 13D/G stake filings — read live from EDGAR full-text search.

Part of Pipeworx — an MCP gateway connecting AI agents to 1856+ live data sources. This is an independent, unofficial integration — not affiliated with, endorsed by, or published by the upstream provider.

Tools

All four take ticker, which also accepts a company name ("NVIDIA Corp") or a bare CIK. insider_trades / insider_8k / insider_13d also take days and limit (max 200).

Counts mean what they say

total_*_filings is the number of rows in filings — the two always agree. The upstream total for the same company and window is reported separately as matched_*_filings, with truncated: true when it is larger. Raise limit to retrieve the rest. (Before 2026-08-31 the count was the upstream total and the rows were a client-side-filtered subset, so the tool could answer {total_form4_filings: 3, filings: []} as a clean 200 — fleet #739.)

A ticker with no matching SEC registrant returns {found: false, reason: "unresolved_company", hint} rather than throwing, so an agent can recover. reason: "missing_ticker" means the argument was empty.

Auth

None. EDGAR is keyless. SEC fair-access rules require a descriptive User-Agent, which the pack sends; requests are subject to SEC's ~10 req/s per-IP limit, so a burst of parallel calls can transiently fail.

Data sources

• EDGAR full-text search — https://efts.sec.gov/LATEST/search-index (forms, ciks, dateRange=custom + startdt/enddt, from for paging)
• SEC company ticker universe — https://www.sec.gov/files/company_tickers.json (ticker/name → CIK, via resolveSecEntity in @pipeworx/shared)
• Filing documents — https://www.sec.gov/Archives/edgar/data/<cik>/… (linked, not fetched)

Upstream gotchas worth knowing

• forms filters the ROOT form. Schedule 13D/G root forms are SCHEDULE 13D / SCHEDULE 13G; the legacy SC 13D spelling matches zero documents and returns a silent empty result.
• efts caches on a key that omits ciks, so a filtered request can be served an unfiltered cached body as a 200. Every request here carries a negated nonsense token (-zqfc<cik>) to make the cache key unique, checks the echoed query.query.bool.filter, and re-checks each row against the CIK.
• dateRange=custom with only startdt is silently ignored — both bounds are always sent.
• display_names carries the ticker inconsistently. An NVDA Form 4 reads NVIDIA CORP (CIK 0001045810) while an NVDA 8-K reads NVIDIA CORP (NVDA) (CIK 0001045810). Never select a company by matching the ticker against that string; filter server-side on ciks.

Quick Start

Add to your MCP client (Claude Desktop, Cursor, Windsurf, etc.):

{
  "mcpServers": {
    "insider-trading": {
      "url": "https://gateway.pipeworx.io/insider-trading/mcp"
    }
  }
}

What this endpoint actually serves

tools/list at https://gateway.pipeworx.io/insider-trading/mcp returns the tools in the table above plus the shared Pipeworx meta-tools — ask_pipeworx, discover_tools, search_within, remember/recall and the rest of the gateway-wide set. So the tool count you see is larger than this table: a single-pack endpoint currently lists roughly 30 shared tools alongside the pack's own. The connection's initialize response states its exact scope, and is the authoritative answer for a given day.

This is deliberate, not multiplexing by accident. The meta-tools are what let a scoped connection answer a question this pack does not cover — via ask_pipeworx, which routes across the whole catalog — without you adding a second MCP server. There is currently no way to mount a pack endpoint without them; if the extra schemas cost you more context than the routing is worth, connect to the full gateway once rather than to several pack endpoints.

Or connect to the full Pipeworx gateway to get every pack's tools listed directly, instead of just this one's:

{
  "mcpServers": {
    "pipeworx": {
      "url": "https://gateway.pipeworx.io/mcp"
    }
  }
}

Both URLs reach the same gateway and the same 1856+ data sources. The only difference is which pack's tools are listed directly; ask_pipeworx reaches all of them from either one.

No MCP client? Call it over HTTP

curl -X POST https://gateway.pipeworx.io/v1/tools/insider_trades \
  -H 'Content-Type: application/json' \
  -d '{"ticker":"NVDA","days":30}'

No account needed for the first calls. Inspect any tool: GET https://gateway.pipeworx.io/v1/tools/insider_trades. Find one: POST https://gateway.pipeworx.io/v1/tools/search_packs with {"query":"..."}.

Run it from npm

@pipeworx/mcp-insider-trading is a local stdio bridge to the Pipeworx gateway: your MCP client starts it with npx, and it forwards every call to the same https://gateway.pipeworx.io/insider-trading/mcp endpoint shown above. It needs network access; there is no offline mode.

{
  "mcpServers": {
    "insider-trading": {
      "command": "npx",
      "args": ["-y", "@pipeworx/mcp-insider-trading"]
    }
  }
}

Without a key it calls anonymously, capped at 50 calls a day. A free account raises that to 200 a day: get a key at https://pipeworx.io/signup and add "env": { "PIPEWORX_API_KEY": "<your key>" } to the server entry. The key is sent as a Bearer token, the same way a hosted connection sends it, so calls are metered and billed exactly as they are there.

It lists only this pack's tools by default. Set PIPEWORX_META_TOOLS=1 to also list ask_pipeworx and the other gateway-wide tools.

Behind an HTTPS proxy: Node's built-in fetch does not read HTTPS_PROXY by default. Newer Node releases honour it when NODE_USE_ENV_PROXY=1 is set; check your Node version's documentation.

Using with ask_pipeworx

Instead of calling tools directly, you can ask questions in plain English — this works on the pack endpoint above as well as on the full gateway:

ask_pipeworx({ question: "your question about Insider Trading data" })

The gateway picks the right tool and fills the arguments automatically.

More

• Docs and guides
• pipeworx.io

License

MIT