@pipeworx/mcp-abn-lookup 0.1.0 → 0.1.1

package/README.md

@@ -4,7 +4,7 @@ BYOK. Australian Business Register lookup: resolve an ABN or ACN to the register
 legal entity name, entity type, ABN status, GST registration, address state/postcode,
 and business names, or search the register by entity/business name.
 
-Part of [Pipeworx](https://pipeworx.io) — an MCP gateway connecting AI agents to 1558+ live data sources.
+Part of [Pipeworx](https://pipeworx.io) — an MCP gateway connecting AI agents to 1679+ live data sources.
 
 ## Tools
 
@@ -97,10 +97,14 @@ directly, instead of just this one's:
 }
 ```
 
-Both URLs reach the same gateway and the same 1558+ data sources. The
+Both URLs reach the same gateway and the same 1679+ 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
+
+This pack takes your own API key (`_apiKey`) — we don't front one for it, so there's no curl here that would run without it. Inspect any tool: `GET https://gateway.pipeworx.io/v1/tools/abn_lookup`. 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

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pipeworx/mcp-abn-lookup",
-  "version": "0.1.0",
+  "version": "0.1.1",
   "description": "ABN Lookup MCP — BYOK wrapper around the Australian Business Register's",
   "type": "module",
   "main": "src/index.ts",
@@ -22,6 +22,11 @@
     "tsx": "^4.19.0"
   },
   "devDependencies": {
-    "typescript": "^5.7.0"
+    "typescript": "^5.9.3",
+    "@cloudflare/workers-types": "^4.20260405.1"
+  },
+  "pipeworx": {
+    "sourceHash": "v1-ea96701fd499288b722815e689fbcdabee808f2665df1a55dcc9a2d9281b6e1f",
+    "sourceCommit": "60f04b68c347e970c6be94a53439d1364061fc06"
   }
 }

package/server.json

@@ -3,7 +3,7 @@
   "name": "io.github.pipeworx-io/abn-lookup",
   "title": "Abn Lookup",
   "description": "ABN Lookup MCP — BYOK wrapper around the Australian Business Register's",
-  "version": "0.1.0",
+  "version": "0.1.1",
   "websiteUrl": "https://pipeworx.io/packs/abn-lookup",
   "repository": {
     "url": "https://github.com/pipeworx-io/mcp-abn-lookup",

package/src/index.ts

@@ -1,6 +1,10 @@
 interface McpToolDefinition {
   name: string;
   description: string;
+  /** Human-facing one-liner (fleet #1967). Optional; consumers fall back to
+   *  description. Kept in step with shared/src/types.ts — scripts/lib/
+   *  check-inlined-types.mjs reports drift at publish time. */
+  summary?: string;
   inputSchema: {
     type: 'object';
     properties: Record<string, unknown>;
@@ -20,6 +24,203 @@ interface McpToolExport {
   provider?: string;
 }
 
+/**
+ * Was this failure OUR OWN web service? — the other half of `internal-db-class.ts`.
+ *
+ * fleet #1089 pulled failures from our own Postgres out of `upstream_down` by
+ * keying on the SQLSTATE inside PostgREST's four-key error envelope. That
+ * covered the majority and structurally could not cover the rest: the rest
+ * never reach Postgres, so they carry no SQLSTATE. What was left, measured over
+ * the 24h to 2026-09-02T15:00Z (fleet #1096):
+ *
+ *     5  pipeworx-catalog  get_pack_tools     Pipeworx catalog error: 522 — error code: 522
+ *     3  fleet             fleet_list_open …  upstream_down: Fleet task queue did not respond within 25s
+ *
+ * 521/522/523/526 are Cloudflare saying its edge could not reach an ORIGIN, and
+ * in both of those rows the origin is ours — `gateway.pipeworx.io` for the
+ * catalog pack (it self-fetches when the gateway hasn't injected a manifest),
+ * our own Supabase for fleet. There is no third party anywhere in either call.
+ * Same defect as #1089: our own outage filed under `upstream_down`, the one
+ * class that means "the source is unreachable and there is nothing for us to
+ * fix", which is why the problem-tools triage skips it.
+ *
+ * WHY NOT A WORDING RULE. The obvious fix is to match `fleet db error:` and
+ * `Pipeworx catalog error:` in classifyToolError. Each is emitted from exactly
+ * one site today, so it would work today. It would also rot the first time
+ * somebody rewords a label — silently, and in the direction of hiding our own
+ * outage, which is worse than the bug being fixed. Every prose rule in
+ * error-class.ts has needed widening as packs invented new wording (#409/#450/
+ * #584); that history is most of that file's comment budget.
+ *
+ * WHAT THIS KEYS ON INSTEAD: **the host the call actually reached.** A URL's
+ * hostname is a fact about the call, not a guess about its prose. Two
+ * consequences that a pack-level flag could not give us, and the reason the
+ * flag was rejected:
+ *
+ *   - It describes the CALL, not the pack. `govcon-intel` fans out to our own
+ *     Supabase AND to genuine third parties; `court-listener` holds our cache
+ *     in Supabase and fetches courtlistener.com. An `internallyHosted: true` on
+ *     either pack would relabel a real third-party outage as ours — inventing
+ *     work, which is the same class of error in the opposite direction.
+ *   - It covers every future internal pack for free, instead of one declared
+ *     slug at a time.
+ *
+ * WHY IT SURVIVES A REWORD. The marker below is not matched as a literal by two
+ * separate files. `markInternalOrigin()` writes it and `internalHostMetricsClass()`
+ * reads it, both from the single exported `INTERNAL_ORIGIN_MARKER` constant in
+ * this module — so changing the wording changes both sides in the same edit and
+ * cannot desynchronise them. The pack's own label (`fleet db error:`,
+ * `Pipeworx catalog error:`) is not read at all: reword it freely, the class is
+ * unaffected. That is the property `stripClassPrefix` lacked when it drifted
+ * from its own classifier three times and needed a CI gate to hold them
+ * together.
+ *
+ * WHERE THE 5xx TEST LIVES. `markInternalOrigin` is called from the places that
+ * hold the real `Response` — `httpError`/`httpErrorMessage` and the timeout
+ * branch of `fetchWithTimeout` in `shared/src/http.ts` — so "is this an
+ * availability failure" is decided from the actual status code, never re-derived
+ * by scraping a number out of a sentence. A 404 from our own registry for a slug
+ * that does not exist is a caller's bad argument and is deliberately NOT marked.
+ */
+
+/**
+ * OUR OWN web service was unreachable — not an upstream, and never `upstream_down`.
+ *
+ * ONE value, not three, unlike `internal_db_*`. That split existed because a
+ * slow query, an exhausted pool and an unknown SQLSTATE have different owners
+ * and different fixes. Here there is only one story to tell — an origin we run
+ * did not answer the edge — and one owner. A bucket with no distinct owner per
+ * value is decoration; #724 is what happens when a class holds several
+ * situations, and inventing sub-values ahead of a reason to act on them
+ * differently is the same mistake with the sign flipped.
+ *
+ * METRICS ONLY, exactly like PLATFORM_KEY_ERROR_CLASS and the internal_db
+ * values. `classifyToolError` still answers `upstream_down` for the retry and
+ * hint paths, which only care whether retrying or a sibling tool might work —
+ * and it might. Nothing a caller sees or is charged changes here.
+ *
+ * READ SIDE: this value is in BROKEN_TOOL_CLASSES, FAULT_CLASSES and
+ * ALL_ERROR_CLASSES in `workers/registry-api/src/index.ts`. All three, or it
+ * lands on no dashboard — fleet #721 is the warning, where the #719 split
+ * worked on the write side and was invisible for weeks.
+ */
+const INTERNAL_SERVICE_UNREACHABLE_CLASS = 'internal_service_unreachable';
+
+/**
+ * The token that carries "this origin is ours" from the call site to the
+ * classifier.
+ *
+ * Appended to the error message rather than attached to the Error object,
+ * because the object does not survive the trip: 275 packs return `{ error:
+ * string }` instead of throwing, the gateway reads `observedError` as a string,
+ * and the fleet pack rebuilds its error from a captured status + body across a
+ * retry loop. A property on an Error would be dropped by every one of those
+ * paths and the class would work in tests and vanish in production.
+ *
+ * WORDING IS LOAD-BEARING, same rule as labelAge's note in authority.ts. This
+ * string is appended to a pack's thrown Error message (shared/src/http.ts),
+ * and a thrown Error's message is exactly what the gateway hands back to the
+ * caller as `content[0].text` when nothing rewrites it (workers/gateway/src
+ * catches the throw and sets `rawResult.message = stripClassPrefix(error)`,
+ * which does not touch this suffix) — so the original wording,
+ * " [pipeworx-hosted origin — our own service, not a third party]", was not a
+ * theoretical leak: it shipped live on pipeworx-catalog's 522s, 7 times in 6
+ * hours on 2026-09-02 (see tests/golden-internal-service.test.ts), verbatim
+ * naming Pipeworx as the host. check:hosting-claims never caught it because it
+ * did not scan shared/ at all (task #2009). Reworded to describe the
+ * OBSERVATION (the origin did not answer) without a claim about who runs it —
+ * the identical fix labelAge got: drop the possessive, keep the fact.
+ */
+const INTERNAL_ORIGIN_MARKER = ' [origin did not respond — retry before concluding the named source is down]';
+
+/**
+ * Supabase's data plane for a project is `<ref>.supabase.co`, where the ref is
+ * exactly twenty lowercase letters (ours is `pqauisounztsgdgfkhke`).
+ *
+ * Matching the shape rather than listing the ref keeps this correct when we add
+ * a project — `supabaseEnv` on a pack entry already points some packs at a
+ * second one — while still excluding `status.supabase.co`, which is Supabase's
+ * own status page and emphatically not our database. Verified 2026-09-02 by
+ * `grep -rhoE '[a-z0-9-]+\.supabase\.(co|in)' mcps shared workers scripts`: the
+ * only real project ref anywhere in the tree is ours, the rest are doc
+ * placeholders (`abc`, `xyz`, `example`) which this pattern also excludes. Same
+ * finding internal-db-class.ts relies on for the PostgREST envelope being ours
+ * by construction.
+ */
+const SUPABASE_PROJECT_HOST = /^[a-z]{20}\.supabase\.(co|in)$/;
+
+/**
+ * Is this a host WE run?
+ *
+ * Deliberately NOT including `*.workers.dev`: plenty of third-party APIs are
+ * hosted on workers.dev, so the suffix says where something runs and not who
+ * owns it. Every internal call we actually make goes to a `pipeworx.io`
+ * hostname or to our Supabase project, both of which are ownership facts.
+ *
+ * `workers/gateway/src/provenance.ts`'s `OUR_HOSTS` answers the same
+ * question and DOES include `workers.dev` — a documented divergence
+ * (task #2051), not a bug to converge. That list decides what a response may
+ * cite as a data SOURCE, where a false negative (citing our own worker as an
+ * external source) is the hosting-disclosure leak this whole file exists to
+ * prevent, so it errs broad. This one decides who gets BLAMED for a 5xx in
+ * outage metrics read by on-call, where a false positive (crediting our own
+ * infra with a third party's outage) hides the real failure, so it errs
+ * narrow. Same suffix, opposite direction, because they are never called for
+ * the same reason.
+ *
+ * Returns false on anything unparseable rather than throwing — this runs inside
+ * an error path, and an error path that can itself throw turns a diagnosable
+ * failure into a mystery.
+ */
+function isPipeworxOrigin(url: string | URL | undefined | null): boolean {
+  if (!url) return false;
+  let host: string;
+  try {
+    host = new URL(url instanceof URL ? url.href : url).hostname.toLowerCase();
+  } catch {
+    return false;
+  }
+  if (host === 'pipeworx.io' || host.endsWith('.pipeworx.io')) return true;
+  return SUPABASE_PROJECT_HOST.test(host);
+}
+
+/**
+ * Append the marker when this failure was OUR origin failing to answer.
+ *
+ * `status` is the HTTP status when there is one, and omitted for a timeout —
+ * where there is no response at all, and "the origin did not answer" is the
+ * whole observation. Statuses below 500 are left alone: a 404 from our own
+ * registry for a slug that does not exist is the caller's argument, not our
+ * outage, and marking it would put ordinary 404s on the incident dashboard.
+ *
+ * Idempotent, so a message that is wrapped and re-marked on the way up (the
+ * fleet pack's retry loop re-throws through two layers) carries the marker once.
+ */
+function markInternalOrigin(
+  message: string,
+  url: string | URL | undefined | null,
+  status?: number,
+): string {
+  if (status !== undefined && status < 500) return message;
+  if (!isPipeworxOrigin(url)) return message;
+  if (message.includes(INTERNAL_ORIGIN_MARKER)) return message;
+  return message + INTERNAL_ORIGIN_MARKER;
+}
+
+/**
+ * Which blob4 value a failure from our own web services books as, or undefined
+ * if this is not one.
+ *
+ * Ordered AFTER `internalDbMetricsClass` at the call site: a PostgREST envelope
+ * from our own Supabase is a strictly more specific statement about the same
+ * row (which of our services, and why), and the two cannot disagree about
+ * whether the failure is ours.
+ */
+function internalHostMetricsClass(error: string): string | undefined {
+  return error.includes(INTERNAL_ORIGIN_MARKER) ? INTERNAL_SERVICE_UNREACHABLE_CLASS : undefined;
+}
+
+
 /**
  * One place to turn a failed `fetch` into an error a caller can act on.
  *
@@ -434,185 +635,6 @@ function pickMessage(node: unknown, depth: number): string | null {
 function collapse(s: string): string {
   return s.replace(/\s+/g, ' ').trim();
 }
-
-/**
- * Was this failure OUR OWN web service? — the other half of `internal-db-class.ts`.
- *
- * fleet #1089 pulled failures from our own Postgres out of `upstream_down` by
- * keying on the SQLSTATE inside PostgREST's four-key error envelope. That
- * covered the majority and structurally could not cover the rest: the rest
- * never reach Postgres, so they carry no SQLSTATE. What was left, measured over
- * the 24h to 2026-09-02T15:00Z (fleet #1096):
- *
- *     5  pipeworx-catalog  get_pack_tools     Pipeworx catalog error: 522 — error code: 522
- *     3  fleet             fleet_list_open …  upstream_down: Fleet task queue did not respond within 25s
- *
- * 521/522/523/526 are Cloudflare saying its edge could not reach an ORIGIN, and
- * in both of those rows the origin is ours — `gateway.pipeworx.io` for the
- * catalog pack (it self-fetches when the gateway hasn't injected a manifest),
- * our own Supabase for fleet. There is no third party anywhere in either call.
- * Same defect as #1089: our own outage filed under `upstream_down`, the one
- * class that means "the source is unreachable and there is nothing for us to
- * fix", which is why the problem-tools triage skips it.
- *
- * WHY NOT A WORDING RULE. The obvious fix is to match `fleet db error:` and
- * `Pipeworx catalog error:` in classifyToolError. Each is emitted from exactly
- * one site today, so it would work today. It would also rot the first time
- * somebody rewords a label — silently, and in the direction of hiding our own
- * outage, which is worse than the bug being fixed. Every prose rule in
- * error-class.ts has needed widening as packs invented new wording (#409/#450/
- * #584); that history is most of that file's comment budget.
- *
- * WHAT THIS KEYS ON INSTEAD: **the host the call actually reached.** A URL's
- * hostname is a fact about the call, not a guess about its prose. Two
- * consequences that a pack-level flag could not give us, and the reason the
- * flag was rejected:
- *
- *   - It describes the CALL, not the pack. `govcon-intel` fans out to our own
- *     Supabase AND to genuine third parties; `court-listener` holds our cache
- *     in Supabase and fetches courtlistener.com. An `internallyHosted: true` on
- *     either pack would relabel a real third-party outage as ours — inventing
- *     work, which is the same class of error in the opposite direction.
- *   - It covers every future internal pack for free, instead of one declared
- *     slug at a time.
- *
- * WHY IT SURVIVES A REWORD. The marker below is not matched as a literal by two
- * separate files. `markInternalOrigin()` writes it and `internalHostMetricsClass()`
- * reads it, both from the single exported `INTERNAL_ORIGIN_MARKER` constant in
- * this module — so changing the wording changes both sides in the same edit and
- * cannot desynchronise them. The pack's own label (`fleet db error:`,
- * `Pipeworx catalog error:`) is not read at all: reword it freely, the class is
- * unaffected. That is the property `stripClassPrefix` lacked when it drifted
- * from its own classifier three times and needed a CI gate to hold them
- * together.
- *
- * WHERE THE 5xx TEST LIVES. `markInternalOrigin` is called from the places that
- * hold the real `Response` — `httpError`/`httpErrorMessage` and the timeout
- * branch of `fetchWithTimeout` in `shared/src/http.ts` — so "is this an
- * availability failure" is decided from the actual status code, never re-derived
- * by scraping a number out of a sentence. A 404 from our own registry for a slug
- * that does not exist is a caller's bad argument and is deliberately NOT marked.
- */
-
-/**
- * OUR OWN web service was unreachable — not an upstream, and never `upstream_down`.
- *
- * ONE value, not three, unlike `internal_db_*`. That split existed because a
- * slow query, an exhausted pool and an unknown SQLSTATE have different owners
- * and different fixes. Here there is only one story to tell — an origin we run
- * did not answer the edge — and one owner. A bucket with no distinct owner per
- * value is decoration; #724 is what happens when a class holds several
- * situations, and inventing sub-values ahead of a reason to act on them
- * differently is the same mistake with the sign flipped.
- *
- * METRICS ONLY, exactly like PLATFORM_KEY_ERROR_CLASS and the internal_db
- * values. `classifyToolError` still answers `upstream_down` for the retry and
- * hint paths, which only care whether retrying or a sibling tool might work —
- * and it might. Nothing a caller sees or is charged changes here.
- *
- * READ SIDE: this value is in BROKEN_TOOL_CLASSES, FAULT_CLASSES and
- * ALL_ERROR_CLASSES in `workers/registry-api/src/index.ts`. All three, or it
- * lands on no dashboard — fleet #721 is the warning, where the #719 split
- * worked on the write side and was invisible for weeks.
- */
-const INTERNAL_SERVICE_UNREACHABLE_CLASS = 'internal_service_unreachable';
-
-/**
- * The token that carries "this origin is ours" from the call site to the
- * classifier.
- *
- * Appended to the error message rather than attached to the Error object,
- * because the object does not survive the trip: 275 packs return `{ error:
- * string }` instead of throwing, the gateway reads `observedError` as a string,
- * and the fleet pack rebuilds its error from a captured status + body across a
- * retry loop. A property on an Error would be dropped by every one of those
- * paths and the class would work in tests and vanish in production.
- *
- * Written as a sentence rather than a sigil because it is going to be read by
- * whoever gets the error, and "our own service, not a third party" is the
- * single most useful thing to tell them — fetchWithTimeout's own comment
- * (fleet #1047) is about exactly this ambiguity, where blaming a healthy vendor
- * by name sent the next person waiting for an outage that did not exist.
- */
-const INTERNAL_ORIGIN_MARKER = ' [pipeworx-hosted origin — our own service, not a third party]';
-
-/**
- * Supabase's data plane for a project is `<ref>.supabase.co`, where the ref is
- * exactly twenty lowercase letters (ours is `pqauisounztsgdgfkhke`).
- *
- * Matching the shape rather than listing the ref keeps this correct when we add
- * a project — `supabaseEnv` on a pack entry already points some packs at a
- * second one — while still excluding `status.supabase.co`, which is Supabase's
- * own status page and emphatically not our database. Verified 2026-09-02 by
- * `grep -rhoE '[a-z0-9-]+\.supabase\.(co|in)' mcps shared workers scripts`: the
- * only real project ref anywhere in the tree is ours, the rest are doc
- * placeholders (`abc`, `xyz`, `example`) which this pattern also excludes. Same
- * finding internal-db-class.ts relies on for the PostgREST envelope being ours
- * by construction.
- */
-const SUPABASE_PROJECT_HOST = /^[a-z]{20}\.supabase\.(co|in)$/;
-
-/**
- * Is this a host WE run?
- *
- * Deliberately NOT including `*.workers.dev`: plenty of third-party APIs are
- * hosted on workers.dev, so the suffix says where something runs and not who
- * owns it. Every internal call we actually make goes to a `pipeworx.io`
- * hostname or to our Supabase project, both of which are ownership facts.
- *
- * Returns false on anything unparseable rather than throwing — this runs inside
- * an error path, and an error path that can itself throw turns a diagnosable
- * failure into a mystery.
- */
-function isPipeworxOrigin(url: string | URL | undefined | null): boolean {
-  if (!url) return false;
-  let host: string;
-  try {
-    host = new URL(url instanceof URL ? url.href : url).hostname.toLowerCase();
-  } catch {
-    return false;
-  }
-  if (host === 'pipeworx.io' || host.endsWith('.pipeworx.io')) return true;
-  return SUPABASE_PROJECT_HOST.test(host);
-}
-
-/**
- * Append the marker when this failure was OUR origin failing to answer.
- *
- * `status` is the HTTP status when there is one, and omitted for a timeout —
- * where there is no response at all, and "the origin did not answer" is the
- * whole observation. Statuses below 500 are left alone: a 404 from our own
- * registry for a slug that does not exist is the caller's argument, not our
- * outage, and marking it would put ordinary 404s on the incident dashboard.
- *
- * Idempotent, so a message that is wrapped and re-marked on the way up (the
- * fleet pack's retry loop re-throws through two layers) carries the marker once.
- */
-function markInternalOrigin(
-  message: string,
-  url: string | URL | undefined | null,
-  status?: number,
-): string {
-  if (status !== undefined && status < 500) return message;
-  if (!isPipeworxOrigin(url)) return message;
-  if (message.includes(INTERNAL_ORIGIN_MARKER)) return message;
-  return message + INTERNAL_ORIGIN_MARKER;
-}
-
-/**
- * Which blob4 value a failure from our own web services books as, or undefined
- * if this is not one.
- *
- * Ordered AFTER `internalDbMetricsClass` at the call site: a PostgREST envelope
- * from our own Supabase is a strictly more specific statement about the same
- * row (which of our services, and why), and the two cannot disagree about
- * whether the failure is ours.
- */
-function internalHostMetricsClass(error: string): string | undefined {
-  return error.includes(INTERNAL_ORIGIN_MARKER) ? INTERNAL_SERVICE_UNREACHABLE_CLASS : undefined;
-}
-
-
 /**
  * ABN Lookup MCP — BYOK wrapper around the Australian Business Register's
  * ABN Lookup JSON web services (https://abr.business.gov.au/json/).

package/src/server.ts

@@ -9,7 +9,7 @@ import { CallToolRequestSchema, ListToolsRequestSchema } from '@modelcontextprot
 import pack from './index.js';
 
 const server = new Server(
-  { name: '@pipeworx/mcp-abn-lookup', version: '0.1.0' },
+  { name: '@pipeworx/mcp-abn-lookup', version: '0.1.1' },
   { capabilities: { tools: {} } },
 );
 

package/tsconfig.json

@@ -3,9 +3,12 @@
     "target": "ES2022",
     "module": "ESNext",
     "moduleResolution": "bundler",
+    "lib": ["ES2022"],
+    "types": ["@cloudflare/workers-types"],
     "strict": true,
     "esModuleInterop": true,
     "skipLibCheck": true,
+    "resolveJsonModule": true,
     "outDir": "dist",
     "rootDir": "src",
     "declaration": true