Technographics MCP Server

@pipeworx/technographics

What a website is built with, read from the page itself. Keyless.

Part of Pipeworx — an MCP gateway connecting AI agents to 1669+ live data sources.

Tools

• tech_detect(url) — the stack a site loads: ecommerce platform, payments, CMS, JS framework, analytics, marketing automation, chat, ad pixels, CDN, web server, visitor-identification tools. Each detection carries the evidence that produced it.
• tech_detect_coverage() — every technology the detector recognises, by category, and what it cannot see.

The route taken, and the licence reasoning

We wrote our own signatures. We did not use a Wappalyzer ruleset. That is the whole licence story and it is the part a future maintainer will need:

Wappalyzer relicensed off MIT before going commercial and every maintained fork inherited the copyleft. An MIT wrapper does not change the licence of the database it consumes, so "use the MIT one" is not the escape it looks like.

What src/signatures.ts contains instead is our own expression of publicly observable facts — that Shopify sets a _shopify_y cookie, that Cloudflare answers with cf-ray, that Next.js emits __NEXT_DATA__. Those facts are not anyone's property; a particular curated database of them is. No GPL code or data is used, imported, or vendored.

What that costs: ~167 technologies across 29 categories, where a commercial set covers tens of thousands. That trade is stated in every response rather than hidden.

Two honest limits, carried on every response

• JavaScript is not executed. We read the HTML the server returned. Anything a tag manager injects client-side is invisible — and for this category that is a specific, large blind spot, not a rounding error, because tag managers are exactly how marketing tools get installed. A site showing only "Google Tag Manager" may be running a dozen tools underneath it.
• Absence is not evidence of absence. A technology missing from an answer may simply be outside the signature set. tech_detect_coverage() exists so a negative can be checked before it is trusted.

Evidence, not booleans

Every detection returns the string that matched — cookie _shopify_y, response header x-vercel-id: …, resource https://js.stripe.com/v3. A detector whose answers cannot be checked is one people stop trusting the first time it is wrong, and during development this caught a real bug: vercel.com reported Mintlify because the page merely names Mintlify in a customer logo. Patterns now key on what a technology emits, never on a product name in prose.

Fetching a caller-supplied URL, safely

This tool fetches an address the caller chose, which is the one shape that turns a data gateway into someone's proxy into private space. So: http/https only, standard ports only, and private, loopback, link-local (including cloud metadata at 169.254.169.254), CGNAT and multicast addresses are refused. Redirects are followed manually with every hop re-checked — a public URL that 302s to the metadata endpoint is the actual attack, and a first-hop-only check does not stop it. Response bodies are capped at 1.5 MB.

Source

The live website. No third-party API, no key, no vendor.

Quick Start

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

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

What this endpoint actually serves

tools/list at https://gateway.pipeworx.io/technographics/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 1669+ 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/tech_detect \
  -H 'Content-Type: application/json' \
  -d '{"url":"shopify.com"}'

No account needed for the first calls. Inspect any tool: GET https://gateway.pipeworx.io/v1/tools/tech_detect. 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": {
    "technographics": {
      "command": "npx",
      "args": ["-y", "@pipeworx/mcp-technographics"]
    }
  }
}

Or run it directly to confirm it starts:

npx -y @pipeworx/mcp-technographics

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 Technographics data" })

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

More

• Docs and guides
• pipeworx.io

License

MIT