Dtcc Swaps MCP Server

@pipeworx/dtcc-swaps

DTCC's public, real-time swap price dissemination feed (pddata.dtcc.com) — individual swap transactions and their pricing terms, disseminated live under CFTC Part 43 for CFTC-regulated swaps (commodity, credit, equity, FX, interest-rate), SEC-regulated security-based swaps (credit, equity, interest-rate) and Canadian-regulator swaps (credit, equity, interest-rate), plus the index of the underlying intraday/end-of-day bulk dissemination files.

Part of Pipeworx — an MCP gateway connecting AI agents to 1683+ live data sources.

Tools

• dtcc_swaps_recent(regulator, asset_class, limit?) — the most-recently disseminated swap transactions for a regulator/asset-class combo (e.g. CFTC/IR), most-recent first, out of the feed's rolling ~100-row window. Each record carries notional, fixed/floating rate terms, execution/effective/ expiration dates, product identifiers, and actionType.
• dtcc_swaps_dissemination_files(regulator, asset_class, granularity?, limit?) — the index of DTCC's public bulk dissemination files (5-minute "slice" files, or one-per-day "cumulative"/EOD files) for a regulator/asset-class, with each file's row count and public download URL. This lists file pointers; it does not download or unzip them.

Auth

Keyless. No account, key, or click-through terms were found anywhere in the reachable surface (see the spike write-up below) — the feed is a public, unauthenticated JSON API on a public S3 bucket with directory listing enabled.

Data sources

• https://pddata.dtcc.com/ppd/api/general/regionname / .../bucketname — resolves the live public S3 bucket name (kgc0418-tdw-data-0 as of 2026-09-23; the pack calls the bucket directly, the same two values the dashboard SPA itself reads at load).
• https://kgc0418-tdw-data-0.s3.amazonaws.com/dashboard/Ticker.json — backs dtcc_swaps_recent. A flat map of {REGULATOR}_{ASSETCLASS} (live keys: CA_CR, CA_EQ, CA_IR, CFTC_CO, CFTC_CR, CFTC_EQ, CFTC_FX, CFTC_IR, SEC_CR, SEC_EQ, SEC_IR) to an array of the ~100 most-recently-disseminated trade records for that bucket. Continuously updated; this pack always fetches it fresh, never caches/mirrors it.
• https://kgc0418-tdw-data-0.s3.amazonaws.com/dashboard/Slice.json (intraday) and .../Cumulative.json (end-of-day) — back dtcc_swaps_dissemination_files. Same {REGULATOR}_{ASSETCLASS} keying, each entry pointing at a downloadable .zip of CSV trade data for that window/day, with a row count.

Traps for the next person

• These endpoints are undiscoverable from the DTCC site's own HTML/JS source by grepping for hostnames — the Angular SPA at pddata.dtcc.com/ppd/ resolves the bucket name from its own /ppd/api/general/* calls at runtime, and reads the JSON files with a relative path built from that response. The bucket name (kgc0418-tdw-data-0) is not literally present anywhere in the shipped JS bundle. It was found by loading the page in a real browser and reading its network requests, not by grepping the bundle. If DTCC ever rotates the bucket, regionname/bucketname are the two calls to re-probe.
• regulator/asset_class combos are not a fixed enum — CA and SEC never carry FX or CO, only CFTC does. Both tools throw a clear error naming the live keys actually present in that call's response rather than trusting a hardcoded list, since DTCC could add/retire a combo without notice.
• Corrections and cancellations are NOT in-place edits. A correction, cancellation, modification or termination of an earlier disseminated trade shows up as its own NEW record in Ticker.json, with actionType set to CORR / EROR / MODI / TERM instead of NEWT. Treat the feed as an append-only event log. A caller who wants only "new trades" should filter actionType === "NEWT" client-side; this pack returns the raw event stream so that behavior isn't silently baked in.
• Ticker.json is a rolling window (~100 rows per bucket), not history. For anything beyond that, dtcc_swaps_dissemination_files points at the underlying zip files (5-min slices or EOD cumulative), but downloading and unzipping those is deliberately out of scope for this pack — it returns the file index (name, URL, row count, window) so a caller can fetch the ones they actually need.
• No licence/terms page was found. www.dtcc.com/legal.php (the general corporate site) 403s on every fetch attempt (Akamai bot wall), unrelated to the pddata.dtcc.com data feed itself, which is unauthenticated throughout. The basis for proxying this data is that DTCC operates as a CFTC/SEC- registered Swap Data Repository and Part 43 of the CFTC's regulations mandates real-time PUBLIC reporting of swap transaction and pricing data — confirmed live via www.cftc.gov/IndustryOversight/DataRepositories/index.htm. Full probe trail: docs/api-wave-2026-09-23-spike.md (source #3).

Quick Start

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

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

What this endpoint actually serves

tools/list at https://gateway.pipeworx.io/dtcc-swaps/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 1683+ 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/dtcc_swaps_recent \
  -H 'Content-Type: application/json' \
  -d '{"regulator":"CFTC","asset_class":"IR","limit":3}'

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

Standalone (no gateway account)

This package also runs as a local stdio MCP server — no Pipeworx account, no gateway round-trip:

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

Or run it directly to confirm it starts:

npx -y @pipeworx/mcp-dtcc-swaps

It speaks MCP over stdin/stdout and answers initialize/tools/list/tools/call for only this pack's tools — none of the shared meta-tools the gateway connection above adds. Same source, same tools, no ask_pipeworx routing.

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 Dtcc Swaps data" })

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

More

• Docs and guides
• pipeworx.io

License

MIT