Mcp Statfin Fi — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited Mcp Statfin Fi (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Statistics Finland (StatFin) — the full catalogue of Finnish official statistics published by Tilastokeskus, served over its PxWeb API: population, education, employment, consumer prices, housing, health and national accounts.
Part of Pipeworx — an MCP gateway connecting AI agents to 1857+ live data sources. This is an independent, unofficial integration — not affiliated with, endorsed by, or published by the upstream provider.
| Tool | What it returns |
|---|---|
subjects | The StatFin subject tree. path empty lists ~135 subject folders; path: "vkour" lists that folder's tables. Entries with type: "l" are folders, type: "t" are tables. |
table_meta | One table's dimensions and every valid value, when you want to see what a table breaks down by. |
query_table | Actual figures. Give it select in plain English and it resolves the codes itself; give it body and it passes your PxWeb query straight through. |
A table path is <folder>/<id>.px where the id is the bare 4-character table code:
vkour/15ig.px Population aged 15 or over by level of education, gender, age and origin
khi/11xs.px Consumer price indexThe longer statfin_<folder>_pxt_<id>.px form appears throughout older StatFin documentation. It was retired upstream and now returns 400 on every table; this pack rewrites it to the current form, so a caller arriving with the documented-but-dead path still gets data (fleet #497).
select maps a dimension to a value using either the English label or the native code, and query_table resolves both against the table's own metadata:
{ "path": "vkour/15ig.px", "select": { "Information": "tertiary level qualification" } }→ 1672232 Finns aged 15+ hold a tertiary qualification (2025), alongside the matching share, 34.5%.
Dimensions you don't name are filled in for you: the year defaults to the latest published, dimensions StatFin can eliminate are aggregated to their total, and a dimension with an explicit total value (SSS, Total, or an open-ended range starting at the table's own floor such as age 15-) uses it. Every one of those choices comes back in `selection_resolved`, and a dimension with no defensible total is returned in full rather than silently collapsed — a defaulted dimension nobody can see is how a wrong number gets read as a right one.
A value that matches several rows returns all of them rather than picking one, so {"Information": "tertiary level qualification"} yields both the count and the percentage. If it matches more than 20, that is reported as too broad.
Results also carry figures: the json-stat2 value[] array paired with its dimension labels, for results up to 60 cells. That is the readable form — json-stat2's flat array over an implicit cartesian product is not.
Even on the /en/ endpoint the dimension codes are Finnish, and some labels are untranslated. There is no way to guess them, which is what select exists to spare you; pass body yourself and you need table_meta first. For vkour/15ig.px:
| Code | Meaning | Total value |
|---|---|---|
ikaryhma_10_20180101 | Age | 15- (not SSS) |
sukupuoli_9_20180101 | Gender | SSS |
syntypera_101_20180101 | Origin (label renders as Syntyperä) | SSS |
timeperiod_y | Year | — |
contentscode | Measure, e.g. kaste5T8 = population with a tertiary level qualification | — |
Worked call — population aged 15+ with a tertiary qualification, 2024:
{
"path": "vkour/15ig.px",
"body": {
"query": [
{ "code": "ikaryhma_10_20180101", "selection": { "filter": "item", "values": ["15-"] } },
{ "code": "sukupuoli_9_20180101", "selection": { "filter": "item", "values": ["SSS"] } },
{ "code": "syntypera_101_20180101", "selection": { "filter": "item", "values": ["SSS"] } },
{ "code": "timeperiod_y", "selection": { "filter": "item", "values": ["2024"] } },
{ "code": "contentscode", "selection": { "filter": "item", "values": ["kaste5T8"] } }
],
"response": { "format": "json-stat2" }
}
}→ value: [1641336], table updated 2026-06-18.
None. Keyless, no registration, no rate-limit key. Reachable from Cloudflare egress.
https://pxdata.stat.fi/PXWeb/api/v1/en/StatFinAdd to your MCP client (Claude Desktop, Cursor, Windsurf, etc.):
{
"mcpServers": {
"statfin-fi": {
"url": "https://gateway.pipeworx.io/statfin-fi/mcp"
}
}
}tools/list at https://gateway.pipeworx.io/statfin-fi/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 1857+ data sources. The only difference is which pack's tools are listed directly; ask_pipeworx reaches all of them from either one.
curl -X POST https://gateway.pipeworx.io/v1/tools/statfin_fi_subjects \
-H 'Content-Type: application/json' \
-d '{"path":""}'No account needed for the first calls. Inspect any tool: GET https://gateway.pipeworx.io/v1/tools/statfin_fi_subjects. Find one: POST https://gateway.pipeworx.io/v1/tools/search_packs with {"query":"..."}.
@pipeworx/mcp-statfin-fi 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/statfin-fi/mcp endpoint shown above. It needs network access; there is no offline mode.
{
"mcpServers": {
"statfin-fi": {
"command": "npx",
"args": ["-y", "@pipeworx/mcp-statfin-fi"]
}
}
}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.
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 Statfin Fi data" })The gateway picks the right tool and fills the arguments automatically.
MIT
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.