@pipeworx/mcp-rba 0.2.2 → 0.3.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/src/index.ts DELETED
@@ -1,1452 +0,0 @@
1
- interface McpToolDefinition {
2
- name: string;
3
- description: string;
4
- /** Human-facing one-liner (fleet #1967). Optional; consumers fall back to
5
- * description. Kept in step with shared/src/types.ts — scripts/lib/
6
- * check-inlined-types.mjs reports drift at publish time. */
7
- summary?: string;
8
- inputSchema: {
9
- type: 'object';
10
- properties: Record<string, unknown>;
11
- required?: string[];
12
- anyOf?: Array<{ required: string[] }>;
13
- oneOf?: Array<{ required: string[] }>;
14
- allOf?: Array<{ required: string[] }>;
15
- };
16
- outputSchema?: Record<string, unknown>;
17
- }
18
-
19
- interface McpToolExport {
20
- tools: McpToolDefinition[];
21
- callTool: (name: string, args: Record<string, unknown>) => Promise<unknown>;
22
- meter?: { credits: number };
23
- cost?: Record<string, unknown>;
24
- provider?: string;
25
- }
26
-
27
- /**
28
- * Was this failure OUR OWN web service? — the other half of `internal-db-class.ts`.
29
- *
30
- * fleet #1089 pulled failures from our own Postgres out of `upstream_down` by
31
- * keying on the SQLSTATE inside PostgREST's four-key error envelope. That
32
- * covered the majority and structurally could not cover the rest: the rest
33
- * never reach Postgres, so they carry no SQLSTATE. What was left, measured over
34
- * the 24h to 2026-09-02T15:00Z (fleet #1096):
35
- *
36
- * 5 pipeworx-catalog get_pack_tools Pipeworx catalog error: 522 — error code: 522
37
- * 3 fleet fleet_list_open … upstream_down: Fleet task queue did not respond within 25s
38
- *
39
- * 521/522/523/526 are Cloudflare saying its edge could not reach an ORIGIN, and
40
- * in both of those rows the origin is ours — `gateway.pipeworx.io` for the
41
- * catalog pack (it self-fetches when the gateway hasn't injected a manifest),
42
- * our own Supabase for fleet. There is no third party anywhere in either call.
43
- * Same defect as #1089: our own outage filed under `upstream_down`, the one
44
- * class that means "the source is unreachable and there is nothing for us to
45
- * fix", which is why the problem-tools triage skips it.
46
- *
47
- * WHY NOT A WORDING RULE. The obvious fix is to match `fleet db error:` and
48
- * `Pipeworx catalog error:` in classifyToolError. Each is emitted from exactly
49
- * one site today, so it would work today. It would also rot the first time
50
- * somebody rewords a label — silently, and in the direction of hiding our own
51
- * outage, which is worse than the bug being fixed. Every prose rule in
52
- * error-class.ts has needed widening as packs invented new wording (#409/#450/
53
- * #584); that history is most of that file's comment budget.
54
- *
55
- * WHAT THIS KEYS ON INSTEAD: **the host the call actually reached.** A URL's
56
- * hostname is a fact about the call, not a guess about its prose. Two
57
- * consequences that a pack-level flag could not give us, and the reason the
58
- * flag was rejected:
59
- *
60
- * - It describes the CALL, not the pack. `govcon-intel` fans out to our own
61
- * Supabase AND to genuine third parties; `court-listener` holds our cache
62
- * in Supabase and fetches courtlistener.com. An `internallyHosted: true` on
63
- * either pack would relabel a real third-party outage as ours — inventing
64
- * work, which is the same class of error in the opposite direction.
65
- * - It covers every future internal pack for free, instead of one declared
66
- * slug at a time.
67
- *
68
- * WHY IT SURVIVES A REWORD. The marker below is not matched as a literal by two
69
- * separate files. `markInternalOrigin()` writes it and `internalHostMetricsClass()`
70
- * reads it, both from the single exported `INTERNAL_ORIGIN_MARKER` constant in
71
- * this module — so changing the wording changes both sides in the same edit and
72
- * cannot desynchronise them. The pack's own label (`fleet db error:`,
73
- * `Pipeworx catalog error:`) is not read at all: reword it freely, the class is
74
- * unaffected. That is the property `stripClassPrefix` lacked when it drifted
75
- * from its own classifier three times and needed a CI gate to hold them
76
- * together.
77
- *
78
- * WHERE THE 5xx TEST LIVES. `markInternalOrigin` is called from the places that
79
- * hold the real `Response` — `httpError`/`httpErrorMessage` and the timeout
80
- * branch of `fetchWithTimeout` in `shared/src/http.ts` — so "is this an
81
- * availability failure" is decided from the actual status code, never re-derived
82
- * by scraping a number out of a sentence. A 404 from our own registry for a slug
83
- * that does not exist is a caller's bad argument and is deliberately NOT marked.
84
- */
85
-
86
- /**
87
- * OUR OWN web service was unreachable — not an upstream, and never `upstream_down`.
88
- *
89
- * ONE value, not three, unlike `internal_db_*`. That split existed because a
90
- * slow query, an exhausted pool and an unknown SQLSTATE have different owners
91
- * and different fixes. Here there is only one story to tell — an origin we run
92
- * did not answer the edge — and one owner. A bucket with no distinct owner per
93
- * value is decoration; #724 is what happens when a class holds several
94
- * situations, and inventing sub-values ahead of a reason to act on them
95
- * differently is the same mistake with the sign flipped.
96
- *
97
- * METRICS ONLY, exactly like PLATFORM_KEY_ERROR_CLASS and the internal_db
98
- * values. `classifyToolError` still answers `upstream_down` for the retry and
99
- * hint paths, which only care whether retrying or a sibling tool might work —
100
- * and it might. Nothing a caller sees or is charged changes here.
101
- *
102
- * READ SIDE: this value is in BROKEN_TOOL_CLASSES, FAULT_CLASSES and
103
- * ALL_ERROR_CLASSES in `workers/registry-api/src/index.ts`. All three, or it
104
- * lands on no dashboard — fleet #721 is the warning, where the #719 split
105
- * worked on the write side and was invisible for weeks.
106
- */
107
- const INTERNAL_SERVICE_UNREACHABLE_CLASS = 'internal_service_unreachable';
108
-
109
- /**
110
- * The token that carries "this origin is ours" from the call site to the
111
- * classifier.
112
- *
113
- * Appended to the error message rather than attached to the Error object,
114
- * because the object does not survive the trip: 275 packs return `{ error:
115
- * string }` instead of throwing, the gateway reads `observedError` as a string,
116
- * and the fleet pack rebuilds its error from a captured status + body across a
117
- * retry loop. A property on an Error would be dropped by every one of those
118
- * paths and the class would work in tests and vanish in production.
119
- *
120
- * WORDING IS LOAD-BEARING, same rule as labelAge's note in authority.ts. This
121
- * string is appended to a pack's thrown Error message (shared/src/http.ts),
122
- * and a thrown Error's message is exactly what the gateway hands back to the
123
- * caller as `content[0].text` when nothing rewrites it (workers/gateway/src
124
- * catches the throw and sets `rawResult.message = stripClassPrefix(error)`,
125
- * which does not touch this suffix) — so the original wording,
126
- * " [pipeworx-hosted origin — our own service, not a third party]", was not a
127
- * theoretical leak: it shipped live on pipeworx-catalog's 522s, 7 times in 6
128
- * hours on 2026-09-02 (see tests/golden-internal-service.test.ts), verbatim
129
- * naming Pipeworx as the host. check:hosting-claims never caught it because it
130
- * did not scan shared/ at all (task #2009). Reworded to describe the
131
- * OBSERVATION (the origin did not answer) without a claim about who runs it —
132
- * the identical fix labelAge got: drop the possessive, keep the fact.
133
- */
134
- const INTERNAL_ORIGIN_MARKER = ' [origin did not respond — retry before concluding the named source is down]';
135
-
136
- /**
137
- * Supabase's data plane for a project is `<ref>.supabase.co`, where the ref is
138
- * exactly twenty lowercase letters (ours is `pqauisounztsgdgfkhke`).
139
- *
140
- * Matching the shape rather than listing the ref keeps this correct when we add
141
- * a project — `supabaseEnv` on a pack entry already points some packs at a
142
- * second one — while still excluding `status.supabase.co`, which is Supabase's
143
- * own status page and emphatically not our database. Verified 2026-09-02 by
144
- * `grep -rhoE '[a-z0-9-]+\.supabase\.(co|in)' mcps shared workers scripts`: the
145
- * only real project ref anywhere in the tree is ours, the rest are doc
146
- * placeholders (`abc`, `xyz`, `example`) which this pattern also excludes. Same
147
- * finding internal-db-class.ts relies on for the PostgREST envelope being ours
148
- * by construction.
149
- */
150
- const SUPABASE_PROJECT_HOST = /^[a-z]{20}\.supabase\.(co|in)$/;
151
-
152
- /**
153
- * Is this a host WE run?
154
- *
155
- * Deliberately NOT including `*.workers.dev`: plenty of third-party APIs are
156
- * hosted on workers.dev, so the suffix says where something runs and not who
157
- * owns it. Every internal call we actually make goes to a `pipeworx.io`
158
- * hostname or to our Supabase project, both of which are ownership facts.
159
- *
160
- * `workers/gateway/src/provenance.ts`'s `OUR_HOSTS` answers the same
161
- * question and DOES include `workers.dev` — a documented divergence
162
- * (task #2051), not a bug to converge. That list decides what a response may
163
- * cite as a data SOURCE, where a false negative (citing our own worker as an
164
- * external source) is the hosting-disclosure leak this whole file exists to
165
- * prevent, so it errs broad. This one decides who gets BLAMED for a 5xx in
166
- * outage metrics read by on-call, where a false positive (crediting our own
167
- * infra with a third party's outage) hides the real failure, so it errs
168
- * narrow. Same suffix, opposite direction, because they are never called for
169
- * the same reason.
170
- *
171
- * Returns false on anything unparseable rather than throwing — this runs inside
172
- * an error path, and an error path that can itself throw turns a diagnosable
173
- * failure into a mystery.
174
- */
175
- function isPipeworxOrigin(url: string | URL | undefined | null): boolean {
176
- if (!url) return false;
177
- let host: string;
178
- try {
179
- host = new URL(url instanceof URL ? url.href : url).hostname.toLowerCase();
180
- } catch {
181
- return false;
182
- }
183
- if (host === 'pipeworx.io' || host.endsWith('.pipeworx.io')) return true;
184
- return SUPABASE_PROJECT_HOST.test(host);
185
- }
186
-
187
- /**
188
- * Append the marker when this failure was OUR origin failing to answer.
189
- *
190
- * `status` is the HTTP status when there is one, and omitted for a timeout —
191
- * where there is no response at all, and "the origin did not answer" is the
192
- * whole observation. Statuses below 500 are left alone: a 404 from our own
193
- * registry for a slug that does not exist is the caller's argument, not our
194
- * outage, and marking it would put ordinary 404s on the incident dashboard.
195
- *
196
- * Idempotent, so a message that is wrapped and re-marked on the way up (the
197
- * fleet pack's retry loop re-throws through two layers) carries the marker once.
198
- */
199
- function markInternalOrigin(
200
- message: string,
201
- url: string | URL | undefined | null,
202
- status?: number,
203
- ): string {
204
- if (status !== undefined && status < 500) return message;
205
- if (!isPipeworxOrigin(url)) return message;
206
- if (message.includes(INTERNAL_ORIGIN_MARKER)) return message;
207
- return message + INTERNAL_ORIGIN_MARKER;
208
- }
209
-
210
- /**
211
- * Which blob4 value a failure from our own web services books as, or undefined
212
- * if this is not one.
213
- *
214
- * Ordered AFTER `internalDbMetricsClass` at the call site: a PostgREST envelope
215
- * from our own Supabase is a strictly more specific statement about the same
216
- * row (which of our services, and why), and the two cannot disagree about
217
- * whether the failure is ours.
218
- */
219
- function internalHostMetricsClass(error: string): string | undefined {
220
- return error.includes(INTERNAL_ORIGIN_MARKER) ? INTERNAL_SERVICE_UNREACHABLE_CLASS : undefined;
221
- }
222
-
223
-
224
- /**
225
- * One place to turn a failed `fetch` into an error a caller can act on.
226
- *
227
- * Nearly every pack was written the same way:
228
- *
229
- * if (!res.ok) throw new Error(`Unsplash: ${res.status}`);
230
- *
231
- * which discards the response body — and the body is usually where the upstream
232
- * says what was actually wrong ("**symbol** not found: GBP", "parameter `year`
233
- * out of range", "unknown taxonomy id"). The caller gets a number, cannot
234
- * self-correct, and retries the same broken call. A 2026-07-31 sweep found this
235
- * shape in 481 of 1,400 packs, 47 of them PLATFORM-keyed.
236
- *
237
- * It also hides bugs one level down. Two of the first three packs audited had a
238
- * second defect that only existed because of this line: unsplash's rate-limit
239
- * branch sat BELOW a catch-all and was unreachable, and bea-gov parsed
240
- * `BEAAPI.Error.APIErrorDescription` below a `!res.ok` throw that made the
241
- * parsing dead code for every non-200.
242
- *
243
- * DELIBERATELY NOT A CLASSIFIER. It does not add `user_error:` /
244
- * `upstream_down:` prefixes. Those decide which tier a failure lands in, and the
245
- * `error` tier is what the daily problem-tools list is built from — it means
246
- * "Pipeworx has a defect". A 400 is genuinely ambiguous: often a caller's bad
247
- * argument, but sometimes a query WE built wrong (ted-eu comma-joined its CPV
248
- * values into something TED rejected, and that bug was found only because it sat
249
- * in `error`). Blanket-classifying 400s as caller mistakes would have hidden it.
250
- * A pack that KNOWS which it is should keep saying so explicitly; this helper is
251
- * for the 481 that say nothing at all.
252
- */
253
-
254
- /** Longest upstream explanation we'll pass through. Enough for a real message,
255
- * short enough that an HTML page or a stack trace can't swamp the error. */
256
-
257
- const MAX_DETAIL = 300;
258
-
259
- /**
260
- * Default bound for `fetchWithTimeout` when a pack doesn't state its own.
261
- *
262
- * 25s mirrors the number `epo-ops` landed on after measuring the real failure:
263
- * a degraded upstream that doesn't error, it just never answers, and a Worker
264
- * sits in `await fetch()` until ITS OWN execution budget kills the request —
265
- * which can take minutes, not seconds (epo_ops_search_patents measured 4-8
266
- * MINUTE hangs before this existed). 25s is short enough that a caller gets a
267
- * fast, actionable error instead of holding the connection, and long enough
268
- * that it doesn't false-trip on a merely-slow-but-alive upstream.
269
- */
270
- const DEFAULT_FETCH_TIMEOUT_MS = 25_000;
271
-
272
- /**
273
- * Read the body of a failed response and fold it into a throwable Error.
274
- *
275
- * Usage — note the `await`, which is the one thing that makes this a mechanical
276
- * change rather than a drop-in:
277
- *
278
- * if (!res.ok) throw await httpError(res, 'Unsplash');
279
- *
280
- * Safe to call on any non-ok response: a body that is missing, empty, unreadable
281
- * or HTML degrades to exactly the old `Name: 404` string rather than throwing
282
- * something new from inside the error path.
283
- */
284
- async function httpError(res: Response, name: string): Promise<Error> {
285
- return new Error(await httpErrorMessage(res, name));
286
- }
287
-
288
- /** The message text without constructing an Error — for packs that need to wrap
289
- * it in their own envelope or add an explicit classification prefix. */
290
- async function httpErrorMessage(res: Response, name: string): Promise<string> {
291
- // The one place a 5xx from a host WE run gets stamped as ours. `res.url` is
292
- // the URL the fetch actually resolved to (after redirects), so this is a fact
293
- // about the call rather than a guess from the `name` the pack passed in —
294
- // reword that label freely, the class does not move. See
295
- // internal-host-class.ts; no-op for every third-party upstream, which is why
296
- // this touches 481 packs' error text and changes none of it.
297
- return markInternalOrigin(
298
- `${name}: ${res.status}${detailSuffix(await readDetail(res))}`,
299
- res.url,
300
- res.status,
301
- );
302
- }
303
-
304
- /**
305
- * Just the upstream's own explanation — no name, no status.
306
- *
307
- * For a pack that has already said both in its own sentence. epo-ops reads
308
- * `EPO rejected this search as too large (HTTP 413) — ${httpErrorMessage(…)}`,
309
- * which rendered as `… (HTTP 413) — EPO: 413.` once the XML detail was being
310
- * dropped: the upstream named twice, the status twice, and the one thing EPO
311
- * actually said ("Not enough characters before truncation character") nowhere
312
- * (fleet #712). Returns '' when the body carries nothing readable, so a caller
313
- * can fall back to its own wording.
314
- */
315
- async function upstreamDetail(res: Response): Promise<string> {
316
- return readDetail(res);
317
- }
318
-
319
- /**
320
- * Read a SUCCESSFUL response as JSON, failing loudly when it isn't JSON.
321
- *
322
- * `httpError` above only ever runs on `!res.ok`, which leaves the nastier half
323
- * of the problem unhandled: an upstream that answers **HTTP 200 with an HTML
324
- * page**. A bot wall, a login redirect, a maintenance interstitial and a CDN
325
- * error page are all 200s, so `res.ok` is true, and `res.json()` then throws
326
- * `Unexpected token '<', "<!DOCTYPE "... is not valid JSON`.
327
- *
328
- * That string is the problem. It names no upstream, carries no status, and
329
- * reads like a parser bug in Pipeworx — so it lands in the `error` tier, which
330
- * means "we have a defect", and the caller is told nothing they can act on.
331
- * data.govt.nz sat dead behind an Imperva challenge this way and every
332
- * status-code health check we own reported it green (7889a845). A zero-length
333
- * body has the same shape: `Unexpected end of JSON input`, seen this week on
334
- * uk-gazette (83% of external calls) and census.
335
- *
336
- * UNLIKE `httpError`, this one DOES classify, and the asymmetry is deliberate.
337
- * A 400 is genuinely ambiguous — often the caller's bad argument, sometimes a
338
- * query we built wrong — so blanket-classifying it would hide our own bugs.
339
- * There is no such ambiguity here: **no argument a caller can pass makes a JSON
340
- * API return an HTML page.** It is always the upstream, so `upstream_down:` is
341
- * a statement of fact rather than a guess, and it keeps these out of the
342
- * problem-tools list where they crowd out real defects.
343
- *
344
- * const data = await parseJson<Feed>(res, 'UK Gazette');
345
- *
346
- * Call it only after the `!res.ok` check — on a failed response you want
347
- * `httpError`, which mines the body for the upstream's own explanation.
348
- */
349
- async function parseJson<T>(res: Response, name: string): Promise<T> {
350
- let raw: string;
351
- try {
352
- raw = await res.text();
353
- } catch {
354
- throw new Error(
355
- `upstream_down: ${name} returned a body that could not be read (HTTP ${res.status}). ` +
356
- 'The connection most likely dropped mid-response; retrying is reasonable.',
357
- );
358
- }
359
-
360
- const type = res.headers.get('content-type') ?? 'no content-type';
361
-
362
- if (!raw.trim()) {
363
- throw new Error(
364
- `upstream_down: ${name} answered HTTP ${res.status} with an EMPTY body where JSON was expected (${type}). ` +
365
- 'Nothing about the request can cause this — it is an upstream fault, and the same call may well work on retry.',
366
- );
367
- }
368
-
369
- // Checked before parsing rather than in the catch, because knowing it is
370
- // markup is what turns "we failed to parse something" into "they served a
371
- // web page" — the second is diagnosable, the first is not.
372
- const head = raw.slice(0, 200).trimStart().toLowerCase();
373
- if (head.startsWith('<!doctype') || head.startsWith('<html') || head.startsWith('<?xml')) {
374
- const kind = head.startsWith('<?xml') ? 'an XML document' : 'an HTML page';
375
- // The summary, not the source. Pasting the first 120 characters of a web
376
- // page handed the agent `<!DOCTYPE html><html lang="en"…` — the same leak
377
- // this branch exists to describe (fleet #712).
378
- throw new Error(
379
- `upstream_down: ${name} answered HTTP ${res.status} with ${kind} instead of JSON (${type}). ` +
380
- 'That is typically a bot wall, a login redirect or a maintenance page — it is returned as a SUCCESS, ' +
381
- `so status-code health checks read it as fine. No argument change will get past it. ` +
382
- `The page says: ${summarizeErrorBody(raw) || 'nothing readable'}`,
383
- );
384
- }
385
-
386
- try {
387
- return JSON.parse(raw) as T;
388
- } catch {
389
- throw new Error(
390
- `upstream_down: ${name} answered HTTP ${res.status} with a body that is not valid JSON (${type}). ` +
391
- `It begins: ${stripMarkup(raw).slice(0, 120) || '(unreadable)'}`,
392
- );
393
- }
394
- }
395
-
396
- /**
397
- * `fetch`, but bounded — the fix for a systemic gap found 2026-08-30: a grep
398
- * audit of every pack's `mcps/*\/src/index.ts` found 1,339 of ~1,500 call
399
- * `fetch()` with NO timeout guard anywhere in the file. Two of those
400
- * (epo-ops, statcan) were confirmed live-hanging for 4-8 minutes before this
401
- * existed — every unguarded call carries the same risk, just unconfirmed.
402
- *
403
- * Mirrors the `epoFetch` wrapper `mcps/epo-ops/src/index.ts` shipped first:
404
- * bound the request with `AbortSignal.timeout`, and on a timeout/abort throw
405
- * an `upstream_down:` error that names the upstream and the bound rather than
406
- * letting the raw `TimeoutError`/`AbortError` (which names neither) propagate.
407
- * `upstream_down:` is deliberate, same reasoning as `parseJson` above — no
408
- * argument a caller passes can make an upstream hang, so it is always the
409
- * upstream's fault, and marking it that way keeps a slow API off the
410
- * problem-tools list where it would crowd out our own defects.
411
- *
412
- * Usage — a mechanical swap for a bare `fetch(url, init)`:
413
- *
414
- * const res = await fetchWithTimeout(url, init, 'Some API');
415
- *
416
- * Pass `timeoutMs` as a fourth argument to override the default for a pack
417
- * with a known-slower upstream; the label should be the same short name you'd
418
- * pass to `httpError`/`httpErrorMessage` for that call.
419
- */
420
- async function fetchWithTimeout(
421
- url: string | URL,
422
- init: RequestInit = {},
423
- name: string,
424
- timeoutMs: number = DEFAULT_FETCH_TIMEOUT_MS,
425
- ): Promise<Response> {
426
- try {
427
- return await fetch(url, { ...init, signal: AbortSignal.timeout(timeoutMs) });
428
- } catch (err) {
429
- if (err instanceof Error && (err.name === 'TimeoutError' || err.name === 'AbortError')) {
430
- // States the OBSERVATION (no response in N seconds), not a diagnosis.
431
- // "appears to be degraded" is an inference about the vendor that we have
432
- // not checked, and it is wrong in a way that misdirects whoever reads it:
433
- // a timeout from a Worker can equally mean OUR egress is blocked.
434
- //
435
- // Measured today (2026-09-01, fleet #1047): every call to
436
- // mainnet.base.org failed from the x402 facilitator while the identical
437
- // request from a laptop returned 200. Base was entirely healthy; the
438
- // public RPC refuses Cloudflare Worker egress. Had this message fired
439
- // there it would have blamed Base by name, and the next person would have
440
- // waited for a vendor outage to clear that did not exist.
441
- // A timeout has no status to test — there is no response at all — so
442
- // `markInternalOrigin` is called without one: an origin we run that never
443
- // answered is an availability failure by definition. This is the half of
444
- // fleet #1096 with neither a SQLSTATE nor a status code to key on.
445
- throw new Error(
446
- markInternalOrigin(
447
- `upstream_down: ${name} did not respond within ${timeoutMs / 1000}s. ` +
448
- `That can be ${name} being slow or down, or this environment being unable to reach it ` +
449
- `(some hosts refuse datacenter/Worker egress) — retry shortly, and check reachability ` +
450
- `from elsewhere before concluding ${name} is down.`,
451
- url,
452
- ),
453
- );
454
- }
455
- // Fleet #2382. Everything that isn't a timeout/abort here is a genuine
456
- // NETWORK-LEVEL failure — DNS resolution, connection refused, TLS handshake,
457
- // Cloudflare's own "Network connection lost." — meaning `fetch()` itself
458
- // threw and no HTTP response of any kind was ever received. Until this fix
459
- // that raw exception was rethrown VERBATIM: a bare `TypeError: fetch failed`
460
- // (or the Workers-runtime equivalent) names no upstream, carries no class
461
- // token, and reads exactly like a defect in OUR code — because it says
462
- // nothing about the call at all. It landed in `error`, the tier that means
463
- // "Pipeworx has a defect", for every one of the (at the time of writing)
464
- // ~470 packs that call this helper directly with no wrapper of their own.
465
- //
466
- // `dexscreener` hit this independently (fleet #1579) and fixed it with a
467
- // bespoke per-pack try/catch around `fetchWithTimeout`. That fix is correct
468
- // but only covers one pack; every other caller of this shared helper still
469
- // leaked the raw exception. Moving the same fix HERE — the one place that
470
- // already carries the timeout case — covers every pack that uses
471
- // `fetchWithTimeout` without a wrapper, for free, and without widening
472
- // `classifyToolError`'s regex list: the fix is giving the message a proper
473
- // `upstream_down:` token at the point the two facts (no response was ever
474
- // received, and which host we were trying to reach) are actually in hand,
475
- // not teaching the classifier to guess from prose after the fact.
476
- //
477
- // Safe on the same grounds as the timeout branch above: no argument a
478
- // caller passes can make `fetch()` itself throw a connection-level error,
479
- // so this is always an availability failure, never a caller mistake. Same
480
- // `markInternalOrigin` treatment — an origin we run that never answered is
481
- // still ours, not a third party's outage.
482
- const raw = err instanceof Error ? err.message : String(err);
483
- throw new Error(
484
- markInternalOrigin(
485
- `upstream_down: could not reach ${name} at all (${raw.slice(0, 160)}). ` +
486
- `No request reached ${name}, so this says NOTHING about whether the arguments you passed ` +
487
- 'are valid — do not re-check them on the strength of this error. Retry shortly.',
488
- url,
489
- ),
490
- );
491
- }
492
- }
493
-
494
- function detailSuffix(detail: string): string {
495
- return detail ? ` — ${detail}` : '';
496
- }
497
-
498
- async function readDetail(res: Response): Promise<string> {
499
- let raw: string;
500
- try {
501
- raw = await res.text();
502
- } catch {
503
- // Body already consumed, or the connection died mid-read. The status alone
504
- // is still worth throwing — never let the error path throw its own error.
505
- return '';
506
- }
507
- return summarizeErrorBody(raw);
508
- }
509
-
510
- /**
511
- * Turn ANY error body — JSON, HTML, XML or plain text — into one short phrase
512
- * that never contains markup.
513
- *
514
- * This used to just drop an HTML or XML body on the floor, on the reasoning
515
- * that markup crowds out the status. That was half right. Dropping it loses the
516
- * one sentence a caller could have acted on: an `Access Denied` title, an SDMX
517
- * `<message:Error>` text, an OPS fault string. A 2026-08-30 support sweep
518
- * measured 13 of 291 caller-facing error rows carrying a raw page or document
519
- * verbatim, across 11 packs, and in every one of them the useful content —
520
- * "Access Denied", "Invalid country code", "SCRAPE_TIMEOUT" — was in there,
521
- * buried in markup the agent had to parse out of a string (fleet #712).
522
- *
523
- * So: extract the meaning, discard the markup. The output is passed through
524
- * `stripMarkup` unconditionally, which is what lets `check:error-body-leak`
525
- * assert mechanically that no caller-facing message can contain `<?xml`,
526
- * `<!DOCTYPE` or `<html`.
527
- */
528
- function summarizeErrorBody(raw: string): string {
529
- if (!raw || !raw.trim()) return '';
530
-
531
- const head = raw.slice(0, 400).trimStart().toLowerCase();
532
-
533
- // An HTML error page (Cloudflare interstitial, nginx default, a login
534
- // redirect) says what it is in its <title>, and almost nowhere else.
535
- if (head.startsWith('<!doctype') || head.startsWith('<html')) {
536
- const title = htmlTitle(raw);
537
- return title
538
- ? `${title} (upstream returned an HTML error page, not an API response)`
539
- : 'upstream returned an HTML error page, not an API response';
540
- }
541
-
542
- // XML fault documents — EPO OPS, SDMX (`<message:Error>`), SOAP faults. The
543
- // human sentence sits in a child element whose tag name says what it is.
544
- if (head.startsWith('<?xml') || head.startsWith('<')) {
545
- const fault = xmlFaultText(raw);
546
- return fault
547
- ? `${stripMarkup(fault).slice(0, MAX_DETAIL)} (from the upstream's XML error document)`
548
- : 'upstream returned an XML error document with no readable message';
549
- }
550
-
551
- // Most JSON error bodies bury one human sentence among ids and echoed request
552
- // params. Prefer that sentence; fall back to the whole body when the shape is
553
- // unfamiliar, since an unfamiliar shape is exactly when we can least afford to
554
- // guess wrong and show nothing.
555
- const fromJson = messageFromJson(raw);
556
- return stripMarkup(fromJson ?? raw).slice(0, MAX_DETAIL);
557
- }
558
-
559
- /** The `<title>` of an HTML error page, or its first `<h1>` — the two places a
560
- * bot wall, a 502 and an "Access Denied" all state what happened. */
561
- function htmlTitle(raw: string): string | null {
562
- const head = raw.slice(0, 4000);
563
- for (const re of [/<title[^>]*>([\s\S]*?)<\/title>/i, /<h1[^>]*>([\s\S]*?)<\/h1>/i]) {
564
- const m = re.exec(head);
565
- const text = m ? stripMarkup(m[1]) : '';
566
- if (text) return text.slice(0, 160);
567
- }
568
- return null;
569
- }
570
-
571
- /** Tag names that carry the explanation in an XML fault document, namespace
572
- * prefix optional (`<message:Error>`, `<com:Text>`, `<faultstring>`). */
573
- const XML_FAULT_TAG_RE =
574
- /<(?:[A-Za-z0-9_.-]+:)?(?:text|message|description|faultstring|reason|detail|title|errormessage|error)\b[^>]*>([^<]{2,400})</i;
575
-
576
- function xmlFaultText(raw: string): string | null {
577
- const head = raw.slice(0, 8000);
578
- const tagged = XML_FAULT_TAG_RE.exec(head);
579
- if (tagged && tagged[1].trim()) return tagged[1];
580
-
581
- // Nothing conventionally named — take the longest text node instead. A fault
582
- // document with one sentence in an oddly named element is still readable;
583
- // returning nothing at all is not.
584
- let best = '';
585
- for (const m of head.matchAll(/>([^<>]{8,400})</g)) {
586
- const text = m[1].trim();
587
- if (text.length > best.length) best = text;
588
- }
589
- return best || null;
590
- }
591
-
592
- /**
593
- * Remove every tag and stray angle bracket, then collapse whitespace.
594
- *
595
- * Applied to everything on the way out, including the JSON and plain-text
596
- * paths, because an upstream is free to embed markup in a JSON string field —
597
- * and a leak is a leak regardless of which branch produced it.
598
- */
599
- function stripMarkup(s: string): string {
600
- return collapse(decodeEntities(s.replace(/<[^>]*>/g, ' ')).replace(/[<>]/g, ' '));
601
- }
602
-
603
- /** The handful of entities that show up in error-page titles. Decoded AFTER
604
- * tags are stripped and BEFORE the angle-bracket sweep, so `&lt;script&gt;`
605
- * in a title cannot decode into markup that survives — EMBL-EBI's ChEMBL 500
606
- * page renders as `500 Internal Server Error &lt; EMBL-EBI` otherwise. */
607
- function decodeEntities(s: string): string {
608
- return s
609
- .replace(/&(?:amp|#0*38);/gi, '&')
610
- .replace(/&(?:lt|#0*60);/gi, '<')
611
- .replace(/&(?:gt|#0*62);/gi, '>')
612
- .replace(/&(?:quot|#0*34);/gi, '"')
613
- .replace(/&(?:#0*39|apos|#x0*27);/gi, "'")
614
- .replace(/&nbsp;/gi, ' ');
615
- }
616
-
617
- /** The conventional "what went wrong" field, under any of the names upstreams
618
- * actually use. Checked in order; first non-empty string wins. */
619
- const MESSAGE_KEYS = [
620
- 'message', 'error_message', 'errorMessage', 'detail', 'details',
621
- 'description', 'error_description', 'reason', 'title', 'fault',
622
- ];
623
-
624
- function messageFromJson(raw: string): string | null {
625
- let parsed: unknown;
626
- try {
627
- parsed = JSON.parse(raw);
628
- } catch {
629
- return null;
630
- }
631
- return pickMessage(parsed, 0);
632
- }
633
-
634
- function pickMessage(node: unknown, depth: number): string | null {
635
- // Two levels covers `{error: {message}}` and `{errors: [{detail}]}`, the two
636
- // shapes that account for nearly all of them, without walking a large payload.
637
- if (depth > 2 || node == null) return null;
638
-
639
- if (typeof node === 'string') return node.trim() || null;
640
-
641
- if (Array.isArray(node)) {
642
- for (const item of node) {
643
- const found = pickMessage(item, depth + 1);
644
- if (found) return found;
645
- }
646
- return null;
647
- }
648
-
649
- if (typeof node !== 'object') return null;
650
- const obj = node as Record<string, unknown>;
651
-
652
- for (const key of MESSAGE_KEYS) {
653
- const v = obj[key];
654
- if (typeof v === 'string' && v.trim()) return v.trim();
655
- }
656
- // `{error: …}` where error is itself an object or a string — the single most
657
- // common wrapper, so it is worth descending into by name rather than scanning
658
- // every key and risking picking up an echoed request parameter.
659
- for (const key of ['error', 'errors', 'fault', 'Error', 'data']) {
660
- if (key in obj) {
661
- const found = pickMessage(obj[key], depth + 1);
662
- if (found) return found;
663
- }
664
- }
665
- return null;
666
- }
667
-
668
- /** Errors are read in a single line of log output; newlines and runs of
669
- * whitespace make a multi-line body unreadable there. */
670
- function collapse(s: string): string {
671
- return s.replace(/\s+/g, ' ').trim();
672
- }
673
- /**
674
- * RBA MCP — Reserve Bank of Australia statistics (free, no auth).
675
- *
676
- * Australia's central bank publishes its statistical tables as CSV at
677
- * rba.gov.au/statistics/tables/csv/<table>-data.csv (keyless). This pack
678
- * parses them: the headline cash rate target, AUD exchange rates, and a
679
- * generic fetch for any RBA table/series. Fills the Australia central-bank
680
- * gap (we had no RBA / Australian monetary-policy source).
681
- *
682
- * Tools:
683
- * - rba_cash_rate: RBA cash rate target (the AU monetary-policy rate)
684
- * - rba_exchange_rates: latest AUD exchange rates vs major currencies
685
- * - rba_series: any RBA statistical series by table id + series id
686
- * - rba_list_series: directory/search over the 71 tables + their series —
687
- * the discovery path rba_series needs and previously
688
- * lacked (fleet #612)
689
- *
690
- * No official RBA endpoint lists every table + series in one call, so
691
- * `RBA_TABLES` below is scraped from rba.gov.au/statistics/tables/ (the
692
- * "Statistical tables" index page groups every table under an <h2> category
693
- * with its id, e.g. "Housing Lending Rates – F6"). Regenerate by re-running
694
- * that scrape if RBA adds/renames tables — it changes rarely.
695
- */
696
-
697
-
698
- // Bound every fetch() in this pack to a fixed timeout — an upstream that
699
- // degrades without erroring would otherwise hold the Worker in `await fetch()`
700
- // until its own execution budget kills the request (minutes, not seconds).
701
- // Mirrors the epoFetch / usaspending retryFetch pattern (fleet #685).
702
- async function pwFetch(url: string | URL, init?: RequestInit): Promise<Response> {
703
- return fetchWithTimeout(url, init ?? {}, 'RBA');
704
- }
705
-
706
- const CSV_BASE = 'https://www.rba.gov.au/statistics/tables/csv';
707
-
708
- interface RbaTableInfo { id: string; title: string; category: string; }
709
-
710
- // Scraped 2026-08-28 from https://www.rba.gov.au/statistics/tables/ — every
711
- // table id currently listed there, with its title and category heading.
712
- const RBA_TABLES: RbaTableInfo[] = [
713
- { id: 'a1', title: 'RBA Balance Sheet - A1', category: 'Reserve Bank of Australia' },
714
- { id: 'a2', title: 'Changes in Monetary Policy and Administered Rates - A2', category: 'Reserve Bank of Australia' },
715
- { id: 'a4', title: 'Foreign Exchange Transactions and Holdings of Official Reserve Assets - A4', category: 'Reserve Bank of Australia' },
716
- { id: 'a5', title: 'Daily Foreign Exchange Market Intervention Transactions - A5', category: 'Reserve Bank of Australia' },
717
- { id: 'a7', title: 'Detected Australian Counterfeits by Denomination - A7', category: 'Reserve Bank of Australia' },
718
- { id: 'b1', title: 'Assets of Financial Institutions - B1', category: 'Assets and Liabilities' },
719
- { id: 'b2', title: 'Banks - Off-balance Sheet Business - B2', category: 'Assets and Liabilities' },
720
- { id: 'b10', title: 'Finance Companies & General Financiers - Selected Assets & Liabilities - B10', category: 'Assets and Liabilities' },
721
- { id: 'b12.1', title: 'Assets of Australian-located Operations - B12.1', category: 'Assets and Liabilities' },
722
- { id: 'b12.2', title: 'Liabilities of Australian-located Operations - B12.2', category: 'Assets and Liabilities' },
723
- { id: 'b13.1', title: 'Consolidated Exposures - Immediate Risk Basis - B13.1', category: 'Assets and Liabilities' },
724
- { id: 'b13.2', title: 'Consolidated Exposures - Ultimate Risk Basis - B13.2', category: 'Assets and Liabilities' },
725
- { id: 'b20', title: 'Renminbi-denominated Deposits and Loans of Australian-resident Financial Institutions - B20', category: 'Assets and Liabilities' },
726
- { id: 'c1', title: 'Credit and Charge Cards – Seasonally Adjusted Series - C1', category: 'Payments System' },
727
- { id: 'c1.2', title: 'Credit and Charge Cards – Original Series – Personal and Commercial Cards - C1.2', category: 'Payments System' },
728
- { id: 'c2', title: 'Debit Cards – Seasonally Adjusted Series - C2', category: 'Payments System' },
729
- { id: 'c3', title: 'Average Merchant Fees for Debit, Credit and Charge Cards - C3', category: 'Payments System' },
730
- { id: 'c4', title: 'ATMs – Seasonally Adjusted Series - C4', category: 'Payments System' },
731
- { id: 'c4.1', title: 'ATMs – Original Series - C4.1', category: 'Payments System' },
732
- { id: 'c5', title: 'Cheques – Seasonally Adjusted Series - C5', category: 'Payments System' },
733
- { id: 'c5.1', title: 'Cheques – Original Series - C5.1', category: 'Payments System' },
734
- { id: 'c6', title: 'Direct Entry and NPP – Seasonally Adjusted Series - C6', category: 'Payments System' },
735
- { id: 'c6.1', title: 'Direct Entry and NPP – Original Series - C6.1', category: 'Payments System' },
736
- { id: 'c7', title: 'Real-time Gross Settlement Statistics - C7', category: 'Payments System' },
737
- { id: 'c9', title: 'Domestic Banking Fees Charged - C9', category: 'Payments System' },
738
- { id: 'd1', title: 'Growth in Selected Financial Aggregates - D1', category: 'Money and Credit Statistics' },
739
- { id: 'd2', title: 'Lending and Credit Aggregates - D2', category: 'Money and Credit Statistics' },
740
- { id: 'd3', title: 'Monetary Aggregates - D3', category: 'Money and Credit Statistics' },
741
- { id: 'd4', title: 'Debt Securities Outstanding - D4', category: 'Money and Credit Statistics' },
742
- { id: 'd5', title: 'Bank Lending Classified by Sector - D5', category: 'Money and Credit Statistics' },
743
- { id: 'd9', title: 'Rural Debt by Lender - D9', category: 'Money and Credit Statistics' },
744
- { id: 'd10', title: 'Margin Lending - D10', category: 'Money and Credit Statistics' },
745
- { id: 'd11', title: 'Monetary Statistics (IMF Framework) - Central Bank Survey - D11', category: 'Money and Credit Statistics' },
746
- { id: 'd12', title: 'Monetary Statistics (IMF Framework) - Authorised Deposit-taking Institutions Survey - D12', category: 'Money and Credit Statistics' },
747
- { id: 'd13', title: 'Monetary Statistics (IMF Framework) - Consolidated Survey - D13', category: 'Money and Credit Statistics' },
748
- { id: 'd14', title: 'Lending to Business - Business Finance Outstanding by Business Size and Interest Rate Type - D14', category: 'Money and Credit Statistics' },
749
- { id: 'd14.1', title: 'Lending to Business - Business Finance Outstanding by Business Size and Industry - D14.1', category: 'Money and Credit Statistics' },
750
- { id: 'e1', title: 'Household and Business Balance Sheets - E1', category: 'Household and Business Finances' },
751
- { id: 'e2', title: 'Household Finances - Selected Ratios - E2', category: 'Household and Business Finances' },
752
- { id: 'e13', title: 'Housing Loan Payments - E13', category: 'Household and Business Finances' },
753
- { id: 'f1', title: 'Interest Rates and Yields - Money Market - Daily - F1', category: 'Interest Rates' },
754
- { id: 'f1.1', title: 'Interest Rates and Yields - Money Market - Monthly - F1.1', category: 'Interest Rates' },
755
- { id: 'f2', title: 'Capital Market Yields - Government Bonds - Daily - F2', category: 'Interest Rates' },
756
- { id: 'f2.1', title: 'Capital Market Yields - Government Bonds - Monthly - F2.1', category: 'Interest Rates' },
757
- { id: 'f3', title: 'Aggregate Measures of Australian Corporate Bond Yields - F3', category: 'Interest Rates' },
758
- { id: 'f4', title: 'Advertised Deposit Rates - F4', category: 'Interest Rates' },
759
- { id: 'f4.1', title: 'Paid Deposit Rates - F4.1', category: 'Interest Rates' },
760
- { id: 'f5', title: 'Indicator Lending Rates - F5', category: 'Interest Rates' },
761
- { id: 'f6', title: 'Housing Lending Rates - F6', category: 'Interest Rates' },
762
- { id: 'f7', title: 'Business Lending Rates - F7', category: 'Interest Rates' },
763
- { id: 'f8', title: 'Personal Lending Rates - F8', category: 'Interest Rates' },
764
- { id: 'f16', title: 'Indicative Mid Rates of Selected Australian Government Securities - F16', category: 'Interest Rates' },
765
- { id: 'f11.1', title: 'Exchange Rates - Daily - 2023 to Current - F11.1', category: 'Exchange Rates' },
766
- { id: 'f9', title: 'Foreign Exchange Turnover Against Australian Dollars - F9', category: 'Exchange Rates' },
767
- { id: 'f10', title: 'Foreign Exchange Turnover Against All Currencies - F10', category: 'Exchange Rates' },
768
- { id: 'f15', title: 'Real Exchange Rate Measures - F15', category: 'Exchange Rates' },
769
- { id: 'f12', title: 'US Dollar Exchange Rates - F12', category: 'Exchange Rates' },
770
- { id: 'g1', title: 'Consumer Price Inflation - G1', category: 'Inflation and Inflation Expectations' },
771
- { id: 'g2', title: 'Consumer Price Inflation - Expenditure Groups - G2', category: 'Inflation and Inflation Expectations' },
772
- { id: 'g3', title: 'Inflation Expectations - G3', category: 'Inflation and Inflation Expectations' },
773
- { id: 'g4', title: 'Consumer Price Inflation - Monthly Collection - G4', category: 'Inflation and Inflation Expectations' },
774
- { id: 'h1', title: 'Gross Domestic Product and Income - H1', category: 'Output and Labour' },
775
- { id: 'h2', title: 'Demand and Income - H2', category: 'Output and Labour' },
776
- { id: 'h3', title: 'Monthly Activity Indicators - H3', category: 'Output and Labour' },
777
- { id: 'h4', title: 'Labour Costs and Productivity - H4', category: 'Output and Labour' },
778
- { id: 'h5', title: 'Labour Force - H5', category: 'Output and Labour' },
779
- { id: 'i1', title: 'International Trade and Balance of Payments - I1', category: 'International Trade and External Finance' },
780
- { id: 'i2', title: 'Commodity Prices - I2', category: 'International Trade and External Finance' },
781
- { id: 'i3', title: 'Balance of Payments - Financial Account - I3', category: 'International Trade and External Finance' },
782
- { id: 'i4', title: 'Australia\'s Gross Foreign Assets and Liabilities - I4', category: 'International Trade and External Finance' },
783
- { id: 'i5', title: 'Australia\'s Net Foreign Liabilities - I5', category: 'International Trade and External Finance' },
784
- ];
785
-
786
- const tools: McpToolExport['tools'] = [
787
- {
788
- name: 'rba_cash_rate',
789
- description:
790
- "The Reserve Bank of Australia's official CASH RATE TARGET — Australia's benchmark monetary-policy interest rate (the AU equivalent of the US fed funds rate). PREFER OVER WEB SEARCH for \"what is the RBA cash rate\", \"Australian interest rate\", \"has the RBA cut rates\". Returns the current rate, the exact date it took effect, the previous rate and when that one took effect, plus recent decision history (table A2, one row per actual decision) — not just a monthly-average series to infer a decision date from. Also returns the F1.1 monthly-average history as before.",
791
- inputSchema: {
792
- type: 'object' as const,
793
- properties: {
794
- recent: { type: 'number', description: 'Number of recent monthly observations to return (1-120, default 12).' },
795
- },
796
- required: [],
797
- },
798
- },
799
- {
800
- name: 'rba_exchange_rates',
801
- description:
802
- 'Latest official RBA exchange rates for the Australian dollar (AUD) against major currencies — USD, EUR, GBP, JPY, CNY, NZD, INR, and more (A$1 = X). PREFER OVER WEB SEARCH for "AUD to USD rate", "Australian dollar exchange rate". Returns the most recent published rates; pass a currency code for that pair\'s recent history.',
803
- inputSchema: {
804
- type: 'object' as const,
805
- properties: {
806
- currency: { type: 'string', description: 'Optional 3-letter code (e.g. "USD", "EUR") to get that pair\'s recent history instead of all latest rates.' },
807
- recent: { type: 'number', description: 'When currency is set: number of recent daily observations (1-60, default 10).' },
808
- },
809
- required: [],
810
- },
811
- },
812
- {
813
- name: 'rba_series',
814
- description:
815
- 'Fetch any RBA statistical series — pass a NAME or description in `series` (e.g. "90-day bank bill rate", "3-year government bond yield") and it resolves the table + series id for you by fuzzy-matching series titles/descriptions across the RBA catalog (the interest-rate and bond-yield tables are searched first), the same way rba_list_series does. Returns recent observations — or every observation in a window when you pass start_date/end_date (YYYY-MM-DD) — plus resolved_from/resolved_to showing what the name resolved to, or a `candidates` list if the name is ambiguous — never guesses a made-up id. Or pass table+series_id directly if you already know them exactly (CPI is g1, monetary aggregates d3, cash rate target is FIRMMCRT in f1.1, etc.). Use rba_cash_rate / rba_exchange_rates for those specific common series.',
816
- inputSchema: {
817
- type: 'object' as const,
818
- properties: {
819
- series: { type: 'string', description: 'A series NAME/description to resolve (e.g. "90-day bank bill rate", "3-year government bond yield") OR an exact series id (e.g. "FIRMMBAB90"). Preferred over table+series_id — resolves the id for you and returns resolved_from/resolved_to (or candidates if ambiguous).' },
820
- table: { type: 'string', description: 'RBA table id, e.g. "f1.1" (money market), "f2.1" (bond yields), "f11.1" (FX), "g1" (CPI), "d3" (monetary aggregates). Optional: narrows `series` resolution to this table; required alongside series_id for the exact legacy path.' },
821
- series_id: { type: 'string', description: 'Legacy: exact RBA series id within `table`, e.g. "FIRMMCRT" (cash rate target), "FXRUSD" (A$/USD). Prefer `series`, which accepts a name and resolves this for you.' },
822
- recent: { type: 'number', description: 'Recent observations to return (1-120, default 12). Ignored when start_date/end_date is set, unless given explicitly.' },
823
- start_date: { type: 'string', description: 'Optional YYYY-MM-DD: return only observations on or after this date (e.g. "2025-09-10"). With a date range, every observation in the range is returned (up to 1000), not just the latest `recent`.' },
824
- end_date: { type: 'string', description: 'Optional YYYY-MM-DD: return only observations on or before this date (e.g. "2025-12-31").' },
825
- },
826
- required: [],
827
- },
828
- },
829
- {
830
- name: 'rba_list_series',
831
- description:
832
- 'Directory of the 71 RBA statistical tables and the series inside them — the discovery tool for rba_series. Pass `query` to search series NAMES/descriptions directly (e.g. "90-day bank bill rate", "3-year government bond yield") and get ranked series_id matches across tables in one call — this is what rba_series uses internally to resolve names, so a hit here is a hit there. Pass `keyword` to search table titles/categories instead (e.g. "housing loan", "cash rate", "inflation", "exchange rate") and see every series in the matched tables. Pass a table id directly to list all its series. Pass nothing to browse every table by category. NOT the same as list_tables (a different pack, Swedish population data) — this is rba_ prefixed.',
833
- inputSchema: {
834
- type: 'object' as const,
835
- properties: {
836
- query: { type: 'string', description: 'Search series NAMES/descriptions directly (not just table titles), e.g. "90-day bank bill rate", "3-year government bond yield" — returns ranked (table, series_id) matches in one call.' },
837
- keyword: { type: 'string', description: 'Optional keyword to search table titles/categories and, for matched tables, their series names — e.g. "housing loan", "cash rate", "CPI", "exchange rate", "credit card".' },
838
- table: { type: 'string', description: 'Optional RBA table id to list every series inside it directly, e.g. "f6" (housing lending rates).' },
839
- },
840
- required: [],
841
- },
842
- },
843
- ];
844
-
845
- // ── CSV parsing ──────────────────────────────────────────────────────
846
- // RBA tables share a fixed shape: a title line, then labelled header rows
847
- // (Title, Description, Frequency, Type, Units, Source, Publication date,
848
- // Series ID), then data rows keyed by DD/MM/YYYY. Columns 1..N are series.
849
-
850
- interface RbaSeries {
851
- series_id: string;
852
- name: string;
853
- description: string;
854
- units: string;
855
- col: number;
856
- }
857
-
858
- function splitCsvLine(line: string): string[] {
859
- const out: string[] = [];
860
- let cur = '';
861
- let q = false;
862
- for (let i = 0; i < line.length; i++) {
863
- const c = line[i];
864
- if (q) {
865
- if (c === '"' && line[i + 1] === '"') { cur += '"'; i++; }
866
- else if (c === '"') q = false;
867
- else cur += c;
868
- } else if (c === '"') q = true;
869
- else if (c === ',') { out.push(cur); cur = ''; }
870
- else cur += c;
871
- }
872
- out.push(cur);
873
- return out;
874
- }
875
-
876
- // RBA tables use two date formats: DD/MM/YYYY (e.g. f1.1) and DD-Mon-YYYY
877
- // (e.g. f11.1 exchange rates). Recognize and normalize both to YYYY-MM-DD.
878
- const MONTHS: Record<string, string> = {
879
- jan: '01', feb: '02', mar: '03', apr: '04', may: '05', jun: '06',
880
- jul: '07', aug: '08', sep: '09', oct: '10', nov: '11', dec: '12',
881
- };
882
- const DATE_RE = /^(\d{2})[/-]([0-9]{2}|[A-Za-z]{3})[/-](\d{4})$/;
883
-
884
- function isDateCell(s: string): boolean {
885
- return DATE_RE.test(s.trim());
886
- }
887
-
888
- function toIso(d: string): string {
889
- const m = DATE_RE.exec(d.trim());
890
- if (!m) return d.trim();
891
- const mon = /^\d{2}$/.test(m[2]) ? m[2] : (MONTHS[m[2].toLowerCase()] ?? m[2]);
892
- return `${m[3]}-${mon}-${m[1]}`;
893
- }
894
-
895
- interface ParsedTable {
896
- table_title: string;
897
- series: RbaSeries[];
898
- rows: { date: string; values: (number | null)[] }[]; // values[col-1]
899
- }
900
-
901
- async function fetchTable(tableId: string): Promise<ParsedTable> {
902
- const id = tableId.trim().toLowerCase().replace(/[^a-z0-9.]/g, '');
903
- const res = await pwFetch(`${CSV_BASE}/${id}-data.csv`, { headers: { 'User-Agent': 'Pipeworx/1.0 (pipeworx.io)' } });
904
- if (!res.ok) throw new Error(`RBA table "${tableId}" not found (HTTP ${res.status}). Check the id at rba.gov.au/statistics/tables.`);
905
- const text = (await res.text()).replace(/^/, '');
906
- const lines = text.split(/\r?\n/);
907
-
908
- let titleRow: string[] = [], descRow: string[] = [], unitsRow: string[] = [], idRow: string[] = [];
909
- let tableTitle = '';
910
- let dataStart = -1;
911
- for (let i = 0; i < lines.length; i++) {
912
- const cells = splitCsvLine(lines[i]);
913
- const head = cells[0]?.trim();
914
- if (i === 0) tableTitle = head ?? '';
915
- if (head === 'Title') titleRow = cells;
916
- else if (head === 'Description') descRow = cells;
917
- else if (head === 'Units') unitsRow = cells;
918
- else if (head === 'Series ID') { idRow = cells; dataStart = i + 1; }
919
- else if (dataStart >= 0 && isDateCell(head ?? '')) { dataStart = i; break; }
920
- }
921
- if (idRow.length === 0) throw new Error(`RBA table "${tableId}" has an unexpected format (no Series ID row).`);
922
-
923
- const series: RbaSeries[] = [];
924
- for (let c = 1; c < idRow.length; c++) {
925
- const sid = idRow[c]?.trim();
926
- if (sid) {
927
- series.push({
928
- series_id: sid,
929
- name: (titleRow[c] ?? '').trim(),
930
- description: (descRow[c] ?? '').trim(),
931
- units: (unitsRow[c] ?? '').trim(),
932
- col: c,
933
- });
934
- }
935
- }
936
-
937
- const rows: ParsedTable['rows'] = [];
938
- for (let i = dataStart; i < lines.length; i++) {
939
- const cells = splitCsvLine(lines[i]);
940
- if (!isDateCell(cells[0]?.trim() ?? '')) continue;
941
- const values = series.map((s) => {
942
- const v = cells[s.col]?.trim();
943
- if (v == null || v === '') return null;
944
- const n = Number(v);
945
- return Number.isFinite(n) ? n : null;
946
- });
947
- rows.push({ date: toIso(cells[0].trim()), values });
948
- }
949
- return { table_title: tableTitle, series, rows };
950
- }
951
-
952
- // Recent observations for one series within a parsed table (skip blank values).
953
- function seriesObservations(t: ParsedTable, seriesId: string, recent: number) {
954
- const idx = t.series.findIndex((s) => s.series_id.toUpperCase() === seriesId.toUpperCase());
955
- if (idx === -1) return null;
956
- const s = t.series[idx];
957
- const obs = t.rows
958
- .map((r) => ({ date: r.date, value: r.values[idx] }))
959
- .filter((o) => o.value !== null);
960
- return { series: s, observations: obs.slice(-Math.max(1, recent)) };
961
- }
962
-
963
- // ── Tool implementations ─────────────────────────────────────────────
964
-
965
- async function cashRate(recent?: number) {
966
- const n = Math.min(120, Math.max(1, recent ?? 12));
967
- const [t, changes] = await Promise.all([fetchTable('f1.1'), fetchTable('a2')]);
968
- const r = seriesObservations(t, 'FIRMMCRT', n);
969
- if (!r) throw new Error('RBA cash rate series (FIRMMCRT) not found in table f1.1.');
970
- const latest = r.observations[r.observations.length - 1] ?? null;
971
- // FIRMMCRT (table F1.1) is a MONTHLY AVERAGE, not a point-in-time value — a
972
- // mid-month decision blends into that month's figure, so it cannot answer
973
- // "effective since" exactly (e.g. a 30-Sep decision barely moves the
974
- // September average). Table A2 "Changes in Monetary Policy and Administered
975
- // Rates" is "As announced" — one row per actual decision, with its exact
976
- // date and the new target — so the computed fields come from there instead
977
- // (fleet #3206).
978
- const decisions = seriesObservations(changes, 'ARBAMPCNCRT', 999999);
979
- const computed = decisions && decisions.observations.length > 0 ? computeFromDecisions(decisions.observations) : null;
980
- return {
981
- ...(computed ?? {}),
982
- series_id: 'FIRMMCRT',
983
- name: r.series.name,
984
- units: r.series.units,
985
- current_rate: latest ? latest.value : null,
986
- as_of: latest ? latest.date : null,
987
- history: r.observations,
988
- note:
989
- 'rate/effective_since/previous_rate/decision_history are computed from table A2 ("Changes in Monetary ' +
990
- 'Policy and Administered Rates" — one row per actual RBA decision, with its exact date). current_rate/' +
991
- 'as_of/history remain the F1.1 monthly-average series as before, which smooths over the exact decision ' +
992
- 'date and should not be used to answer "when did it last change".',
993
- };
994
- }
995
-
996
- /** {rate, effective_since, previous_rate, previous_rate_effective_since, decision_history, data_as_of} from ascending {date,value} decision rows (already one row per change — no de-duplication needed). */
997
- function computeFromDecisions(obs: { date: string; value: number | null }[]) {
998
- const pts = obs.filter((o) => o.value !== null) as { date: string; value: number }[];
999
- if (pts.length === 0) return null;
1000
- const current = pts[pts.length - 1];
1001
- const previous = pts.length > 1 ? pts[pts.length - 2] : null;
1002
- return {
1003
- rate: current.value,
1004
- unit: 'percent',
1005
- effective_since: current.date,
1006
- previous_rate: previous?.value ?? null,
1007
- previous_rate_effective_since: previous?.date ?? null,
1008
- decision_history: pts
1009
- .slice(-5)
1010
- .reverse()
1011
- .map((p) => ({ rate: p.value, effective_since: p.date })),
1012
- data_as_of: current.date,
1013
- };
1014
- }
1015
-
1016
- async function exchangeRates(currency?: string, recent?: number) {
1017
- const t = await fetchTable('f11.1');
1018
- if (currency && currency.trim()) {
1019
- const code = currency.trim().toUpperCase();
1020
- // Map common code -> RBA series id (FXR + code-ish); fall back to matching the name "A$1=CODE".
1021
- const match = t.series.find((s) => s.name.toUpperCase().endsWith(`=${code}`) || s.units.toUpperCase() === code);
1022
- // `not_found` (not the old bespoke `currency_not_found`) is the classifier's
1023
- // recognized routing token — the gateway reads it at position 0 and books
1024
- // this as user_error rather than the `error` tier (fleet #637). A code that
1025
- // isn't in the closed RBA f11.1 set is the caller's identifier, never ours.
1026
- if (!match) return { error: 'not_found', message: `No AUD/${code} series in RBA f11.1. Try USD, EUR, GBP, JPY, CNY, NZD, INR, etc.` };
1027
- const r = seriesObservations(t, match.series_id, Math.min(60, Math.max(1, recent ?? 10)))!;
1028
- return { pair: `AUD/${code}`, series_id: match.series_id, units: match.units, observations: r.observations };
1029
- }
1030
- // Latest rate for every currency: take the last row with a value per series.
1031
- const rates: Record<string, { value: number; date: string }> = {};
1032
- for (let i = 0; i < t.series.length; i++) {
1033
- const s = t.series[i];
1034
- for (let r = t.rows.length - 1; r >= 0; r--) {
1035
- const v = t.rows[r].values[i];
1036
- if (v !== null) { rates[s.units || s.name] = { value: v, date: t.rows[r].date }; break; }
1037
- }
1038
- }
1039
- return { base: 'AUD', note: 'A$1 = value (i.e. how many units of each currency one Australian dollar buys; TWI/SDR are indices).', latest_rates: rates };
1040
- }
1041
-
1042
- interface DateRange { start?: string; end?: string }
1043
-
1044
- const ISO_DAY = /^\d{4}-\d{2}-\d{2}$/;
1045
- const RANGE_CAP = 1000;
1046
-
1047
- function parseRange(start?: unknown, end?: unknown): DateRange | { error: string; message: string } {
1048
- const out: DateRange = {};
1049
- for (const [k, v] of [['start_date', start], ['end_date', end]] as const) {
1050
- if (v == null || v === '') continue;
1051
- const s = String(v).trim();
1052
- if (!ISO_DAY.test(s)) return { error: 'invalid_argument', message: `${k} must be YYYY-MM-DD, got "${s}".` };
1053
- if (k === 'start_date') out.start = s; else out.end = s;
1054
- }
1055
- if (out.start && out.end && out.start > out.end) {
1056
- return { error: 'invalid_argument', message: `start_date ${out.start} is after end_date ${out.end}.` };
1057
- }
1058
- return out;
1059
- }
1060
-
1061
- async function genericSeries(table: string, seriesId: string, recent?: number, range: DateRange = {}) {
1062
- const ranged = !!(range.start || range.end);
1063
- // A date range asks for every observation in it — the old `recent` default
1064
- // of 12 would silently return the LATEST 12 regardless of the range, which
1065
- // is the #2305 failure (a 2025 range answered with 2026 data, outcome ok).
1066
- const n = ranged && recent == null ? Number.MAX_SAFE_INTEGER : Math.min(120, Math.max(1, recent ?? 12));
1067
- const t = await fetchTable(table);
1068
- const r = seriesObservations(t, seriesId, ranged ? Number.MAX_SAFE_INTEGER : n);
1069
- if (!r) {
1070
- // Deliberately NOT a truncated inline list (fleet #612): the old message
1071
- // sliced to 25 ids and appended a Unicode ellipsis ("…", U+2026) when a
1072
- // table had more — a client that mis-decodes our UTF-8 as Latin-1 turns
1073
- // that single character into a garbled "â¦" tail, which is exactly the
1074
- // shape support saw stuck to a caller's invented series id. Point at the
1075
- // untruncated tool instead of reproducing the list (and its failure mode)
1076
- // inline; keep a short, ellipsis-free sample for a quick eyeball.
1077
- const sample = t.series.slice(0, 8).map((s) => s.series_id).join(', ');
1078
- const rest = t.series.length > 8 ? ` (${t.series.length} series total)` : '';
1079
- // `not_found` (not the old bespoke `series_not_found`) is the classifier's
1080
- // recognized routing token — a series id the caller invented is theirs to
1081
- // fix, not a Pipeworx defect (fleet #637).
1082
- return {
1083
- error: 'not_found',
1084
- message: `Series "${seriesId}" not in RBA table "${table}". Call rba_list_series({table:"${table}"}) for the full list of series ids, names and units in this table. First few: ${sample}${rest}.`,
1085
- };
1086
- }
1087
- let truncated = false;
1088
- if (ranged) {
1089
- const all = r.observations;
1090
- const inRange = all.filter((o) => (!range.start || o.date >= range.start) && (!range.end || o.date <= range.end));
1091
- if (inRange.length === 0) {
1092
- // Never answer an out-of-range question with other dates' data. Say
1093
- // what the table actually covers so the caller can tell "no data in
1094
- // that window" from "this table starts later" (RBA splits history into
1095
- // separate historical tables).
1096
- return {
1097
- error: 'not_found',
1098
- message: `No ${r.series.series_id} observations in RBA table "${table}" between ${range.start ?? 'the start'} and ${range.end ?? 'the end'}. This table covers ${all[0]?.date ?? '?'} to ${all[all.length - 1]?.date ?? '?'}; older history is in the RBA's separate historical tables.`,
1099
- };
1100
- }
1101
- const cap = Math.min(n, RANGE_CAP);
1102
- truncated = inRange.length > cap;
1103
- r.observations = truncated ? inRange.slice(0, cap) : inRange;
1104
- }
1105
- // Say what the answer is ABOUT, in words (fleet #2238). The old payload was
1106
- // a table id, a series id and numbers — nothing a reader (or the gateway's
1107
- // entity check, workers/gateway/src/entity.ts) could recognise as "the
1108
- // Reserve Bank of Australia's AUD/USD series": every rba_series answer
1109
- // scored `unverifiable` (97/97 in the 7 days to 2026-09-18). `country` is a
1110
- // key the detector reads; `statement` is the same fact as prose.
1111
- const first = r.observations[0];
1112
- const last = r.observations[r.observations.length - 1];
1113
- const title = t.table_title.replace(/\s+/g, ' ').trim();
1114
- const statement =
1115
- `Reserve Bank of Australia (RBA) Table ${title}: series ${r.series.series_id} (${r.series.name}), units ${r.series.units}. ` +
1116
- `${r.observations.length} observation(s)` +
1117
- (first && last ? ` from ${first.date} to ${last.date}; latest ${last.value} on ${last.date}` : '') +
1118
- '.';
1119
- return {
1120
- table,
1121
- table_title: t.table_title,
1122
- series_id: r.series.series_id,
1123
- name: r.series.name,
1124
- series_name: r.series.name,
1125
- units: r.series.units,
1126
- source: 'Reserve Bank of Australia',
1127
- country: 'Australia',
1128
- statement,
1129
- ...(ranged ? { start_date: range.start ?? null, end_date: range.end ?? null, truncated } : {}),
1130
- observations: r.observations,
1131
- };
1132
- }
1133
-
1134
- // ── Discovery: rba_list_series (fleet #612) ─────────────────────────
1135
- //
1136
- // RBA_TABLES gives us titles/categories without a network call. Resolving
1137
- // KEYWORD to actual SERIES still needs the matched table's live CSV (only
1138
- // RBA_TABLES has titles; series ids/names live inside each table's data).
1139
- // Capped so a broad keyword can't turn one call into a dozen upstream fetches.
1140
- const MAX_LIVE_FETCHES = 5;
1141
-
1142
- // ── Name resolution (fleet #1219) ───────────────────────────────────
1143
- //
1144
- // rba_series previously required an exact (table, series_id) pair. The
1145
- // router picks this tool correctly for a query like "90-day bank bill rate"
1146
- // but has no real ids to fill in and INVENTS ones (BBS90, GOVBND3 — neither
1147
- // exists; the real ids are FIRMMBAB90 in f1.1 and FCMYGBAG3 in f2.1). This
1148
- // section resolves a NAME/description to a (table, series_id) by fuzzy-
1149
- // matching series titles + descriptions, so the caller never has to guess.
1150
- //
1151
- // Search scope: the "Interest Rates" tables (money-market rates, bond
1152
- // yields) hold the vast majority of series people name directly — cash
1153
- // rate, bank bills, treasury notes, government bond yields — so they are
1154
- // searched FIRST, within the same live-fetch budget as the keyword search
1155
- // below. If nothing scores there, we fall back to the keyword-based table
1156
- // search (title/category overlap) already used by rba_list_series.
1157
- const CORE_SERIES_TABLES = ['f1.1', 'f2.1', 'f1', 'f2', 'f16'];
1158
-
1159
- function tokenize(s: string): string[] {
1160
- return s.toLowerCase().split(/[^a-z0-9]+/).filter(Boolean);
1161
- }
1162
-
1163
- // Finance shorthand: "90-day bank bill" and "3-month BABs/NCDs" name the
1164
- // same instrument. RBA series titles use the month form; callers (and the
1165
- // router) commonly say the day form. Detect "<N>-day"/"<N> day" in the raw
1166
- // query and add the equivalent month count as extra search tokens.
1167
- const DAY_TO_MONTH: Record<number, number> = { 30: 1, 90: 3, 180: 6, 365: 12 };
1168
- function expandDaySynonyms(raw: string): string[] {
1169
- const extra: string[] = [];
1170
- const re = /(\d+)\s*-?\s*day/gi;
1171
- let m: RegExpExecArray | null;
1172
- while ((m = re.exec(raw))) {
1173
- const days = Number(m[1]);
1174
- const months = DAY_TO_MONTH[days] ?? Math.round(days / 30);
1175
- if (months > 0) extra.push(String(months), 'month', 'months');
1176
- }
1177
- return extra;
1178
- }
1179
-
1180
- interface SeriesMatch { table: string; table_title: string; series_id: string; name: string; units: string; score: number; }
1181
-
1182
- // Score one series against the query tokens/phrase. Word-substring match
1183
- // (not whole-word) so "bill" matches "Bills" and "year" matches "years" —
1184
- // the same tolerance rba_list_series already uses for table titles.
1185
- function scoreSeries(queryLower: string, tokens: string[], s: RbaSeries): number {
1186
- const hay = `${s.name} ${s.description}`.toLowerCase();
1187
- if (s.series_id.toLowerCase() === queryLower.replace(/\s+/g, '')) return 1000; // exact id, case-insensitive
1188
- let score = hay.includes(queryLower) ? 20 : 0; // exact phrase bonus
1189
- for (const t of tokens) if (hay.includes(t)) score += 1;
1190
- return score;
1191
- }
1192
-
1193
- async function searchSeriesAcrossTables(query: string, tableHint?: string): Promise<{ matches: SeriesMatch[]; searched_tables: string[] }> {
1194
- const queryLower = query.trim().toLowerCase();
1195
- const tokens = [...new Set([...tokenize(query), ...expandDaySynonyms(query)])];
1196
-
1197
- async function fetchScored(tableId: string): Promise<SeriesMatch[]> {
1198
- try {
1199
- const t = await fetchTable(tableId);
1200
- return t.series
1201
- .map((s) => ({ table: tableId, table_title: t.table_title, series_id: s.series_id, name: s.name, units: s.units, score: scoreSeries(queryLower, tokens, s) }))
1202
- .filter((m) => m.score > 0);
1203
- } catch {
1204
- return [];
1205
- }
1206
- }
1207
-
1208
- if (tableHint && tableHint.trim()) {
1209
- const id = tableHint.trim().toLowerCase();
1210
- const matches = dedupeMatches(await fetchScored(id));
1211
- return { matches, searched_tables: [id] };
1212
- }
1213
-
1214
- // Pass 1: the core interest-rate/bond-yield tables (fits MAX_LIVE_FETCHES exactly).
1215
- const pass1 = (await Promise.all(CORE_SERIES_TABLES.map(fetchScored))).flat();
1216
- const pass1Deduped = dedupeMatches(pass1);
1217
- if (pass1Deduped.length > 0 && pass1Deduped[0].score >= 3) {
1218
- return { matches: pass1Deduped, searched_tables: CORE_SERIES_TABLES };
1219
- }
1220
-
1221
- // Pass 2: broaden using the same title/category keyword scoring rba_list_series
1222
- // uses, excluding tables already tried, capped to the same live-fetch budget.
1223
- const words = tokens;
1224
- const scoredTables = RBA_TABLES
1225
- .filter((t) => !CORE_SERIES_TABLES.includes(t.id))
1226
- .map((t) => {
1227
- const hay = `${t.title} ${t.category}`.toLowerCase();
1228
- let score = hay.includes(queryLower) ? 100 : 0;
1229
- for (const w of words) if (hay.includes(w)) score += 1;
1230
- return { id: t.id, score };
1231
- })
1232
- .filter((x) => x.score > 0)
1233
- .sort((a, b) => b.score - a.score)
1234
- .slice(0, MAX_LIVE_FETCHES)
1235
- .map((x) => x.id);
1236
-
1237
- const pass2 = (await Promise.all(scoredTables.map(fetchScored))).flat();
1238
- const combined = dedupeMatches([...pass1, ...pass2]);
1239
- return { matches: combined, searched_tables: [...CORE_SERIES_TABLES, ...scoredTables] };
1240
- }
1241
-
1242
- // RBA republishes several series in both a daily table (raw/EOD) and a
1243
- // monthly-average table with the same or near-identical name (e.g. "Cash
1244
- // Rate Target" in both f1 and f1.1; "EOD 3-month BABs/NCDs" in f1 vs
1245
- // "3-month BABs/NCDs" in f1.1). Without collapsing these, every named query
1246
- // looks "ambiguous" against its own daily/monthly twin even though a caller
1247
- // asking for "the 90-day bank bill rate" doesn't care which table it lives
1248
- // in. Prefer the CORE_SERIES_TABLES-priority (monthly/interpolated) table
1249
- // as the canonical representative when scores tie.
1250
- function dedupeMatches(matches: SeriesMatch[]): SeriesMatch[] {
1251
- const priority = (tableId: string) => {
1252
- const idx = CORE_SERIES_TABLES.indexOf(tableId);
1253
- return idx === -1 ? CORE_SERIES_TABLES.length : idx;
1254
- };
1255
- const byKey = new Map<string, SeriesMatch>();
1256
- for (const m of matches) {
1257
- const key = m.name.toLowerCase().replace(/^eod\s+/, '').trim();
1258
- const existing = byKey.get(key);
1259
- if (!existing || m.score > existing.score || (m.score === existing.score && priority(m.table) < priority(existing.table))) {
1260
- byKey.set(key, m);
1261
- }
1262
- }
1263
- return [...byKey.values()].sort((a, b) => b.score - a.score);
1264
- }
1265
-
1266
- interface ResolveResult {
1267
- ok: true;
1268
- table: string;
1269
- series_id: string;
1270
- resolved_from: string;
1271
- resolved_to: string;
1272
- }
1273
- interface ResolveAmbiguous {
1274
- ok: false;
1275
- error: 'ambiguous' | 'not_found';
1276
- message: string;
1277
- candidates?: { table: string; series_id: string; name: string; units: string }[];
1278
- }
1279
-
1280
- // A top hit is auto-resolved only when it's clearly ahead of the runner-up —
1281
- // otherwise we hand back candidates rather than guess (the whole point of
1282
- // this fix is to stop inventing ids).
1283
- async function resolveSeriesName(query: string, tableHint?: string): Promise<ResolveResult | ResolveAmbiguous> {
1284
- const { matches, searched_tables } = await searchSeriesAcrossTables(query, tableHint);
1285
- if (matches.length === 0) {
1286
- return {
1287
- ok: false,
1288
- error: 'not_found',
1289
- message: `No RBA series name/description matches "${query}" in tables searched (${searched_tables.join(', ')}). Call rba_list_series({query:"${query}"}) to see what's available, or rba_list_series({keyword:"..."}) to search table titles more broadly.`,
1290
- };
1291
- }
1292
- const top = matches[0];
1293
- const second = matches[1];
1294
- // Daily/monthly twins are already collapsed by dedupeMatches, so a genuine
1295
- // score lead here means the query's own words picked the winner (e.g. the
1296
- // "90-day"→"3 month" synonym separating the 90- from the 180-day bank
1297
- // bill) — any measurable lead is enough; a true tie stays ambiguous.
1298
- const clear = top.score >= 1000 || !second || top.score > second.score;
1299
- if (!clear) {
1300
- return {
1301
- ok: false,
1302
- error: 'ambiguous',
1303
- message: `"${query}" matches multiple RBA series — pass the exact table + series_id, or a more specific name.`,
1304
- candidates: matches.slice(0, 6).map((m) => ({ table: m.table, series_id: m.series_id, name: m.name, units: m.units })),
1305
- };
1306
- }
1307
- return { ok: true, table: top.table, series_id: top.series_id, resolved_from: query, resolved_to: `${top.table}/${top.series_id} (${top.name})` };
1308
- }
1309
-
1310
- async function listSeries(table?: string, keyword?: string, query?: string) {
1311
- if (query && query.trim()) {
1312
- const { matches, searched_tables } = await searchSeriesAcrossTables(query, table);
1313
- return {
1314
- query,
1315
- matched_series: matches.length,
1316
- series: matches.slice(0, 20).map((m) => ({ table: m.table, table_title: m.table_title, series_id: m.series_id, name: m.name, units: m.units, score: m.score })),
1317
- searched_tables,
1318
- ...(matches.length === 0 ? { note: 'No series name/description matched. Try rba_list_series({keyword:"..."}) to search table titles instead, or browse by category with no arguments.' } : {}),
1319
- };
1320
- }
1321
-
1322
- if (table && table.trim()) {
1323
- const t = await fetchTable(table);
1324
- return {
1325
- table: table.trim().toLowerCase(),
1326
- table_title: t.table_title,
1327
- count: t.series.length,
1328
- series: t.series.map((s) => ({ series_id: s.series_id, name: s.name, units: s.units })),
1329
- };
1330
- }
1331
-
1332
- const kw = keyword?.trim().toLowerCase();
1333
- if (kw) {
1334
- // Word-level scoring, not whole-phrase substring: a caller says "housing
1335
- // loan interest rates" and RBA's own title is "Housing Lending Rates" —
1336
- // "loan" vs "lending" never substring-matches, so a naive .includes(kw)
1337
- // finds nothing for exactly the query this tool exists to answer. Score
1338
- // by how many query words appear (in title OR category — F6's category
1339
- // "Interest Rates" is what pulls "interest" and "rates" in), rank by
1340
- // score, and still give an exact whole-phrase hit a strong boost so a
1341
- // precise query doesn't get diluted by broader partial matches.
1342
- const words = kw.split(/\s+/).filter(Boolean);
1343
- const scored = RBA_TABLES
1344
- .map((t) => {
1345
- if (t.id === kw) return { t, score: 1000 };
1346
- const hay = `${t.title} ${t.category}`.toLowerCase();
1347
- let score = hay.includes(kw) ? 100 : 0;
1348
- for (const w of words) if (hay.includes(w)) score += 1;
1349
- return { t, score };
1350
- })
1351
- .filter((x) => x.score > 0)
1352
- .sort((a, b) => b.score - a.score);
1353
- const matches = scored.map((x) => x.t);
1354
- if (matches.length === 0) {
1355
- return {
1356
- found: false,
1357
- message: `No RBA table title or category matches "${keyword}". Categories: ${[...new Set(RBA_TABLES.map((t) => t.category))].join(', ')}. Call rba_list_series with no arguments to browse every table.`,
1358
- };
1359
- }
1360
- const picked = matches.slice(0, MAX_LIVE_FETCHES);
1361
- const tables = await Promise.all(
1362
- picked.map(async (m) => {
1363
- try {
1364
- const t = await fetchTable(m.id);
1365
- return {
1366
- table: m.id,
1367
- table_title: m.title,
1368
- category: m.category,
1369
- series: t.series.map((s) => ({ series_id: s.series_id, name: s.name, units: s.units })),
1370
- };
1371
- } catch {
1372
- // One bad upstream fetch shouldn't sink every other match.
1373
- return {
1374
- table: m.id, table_title: m.title, category: m.category, series: [],
1375
- note: `Could not fetch this table's series right now — try rba_list_series({table:"${m.id}"}) directly.`,
1376
- };
1377
- }
1378
- }),
1379
- );
1380
- return {
1381
- keyword,
1382
- matched_tables: matches.length,
1383
- shown: tables.length,
1384
- ...(matches.length > picked.length ? { not_shown: matches.length - picked.length, note: 'Narrow the keyword to see the rest.' } : {}),
1385
- tables,
1386
- };
1387
- }
1388
-
1389
- // Neither table nor keyword: browse every table by category (no live fetch).
1390
- const byCategory: Record<string, { id: string; title: string }[]> = {};
1391
- for (const t of RBA_TABLES) (byCategory[t.category] ??= []).push({ id: t.id, title: t.title });
1392
- return {
1393
- total_tables: RBA_TABLES.length,
1394
- categories: byCategory,
1395
- note: 'Pass a keyword (e.g. "housing loan", "cash rate", "inflation", "exchange rate") to search titles and see their series, or a table id directly.',
1396
- };
1397
- }
1398
-
1399
- // rba_series entry point: resolves `series` (name or id) to a (table,
1400
- // series_id) pair when the legacy exact table+series_id pair isn't given,
1401
- // then fetches observations. Never invents an id — returns `ambiguous`
1402
- // candidates or `not_found` instead of guessing (fleet #1219).
1403
- async function rbaSeriesTool(table?: string, seriesIdArg?: string, seriesArg?: string, recent?: number, startDate?: unknown, endDate?: unknown) {
1404
- const range = parseRange(startDate, endDate);
1405
- if ('error' in range) return range;
1406
- // Legacy exact path: both table and series_id given — unchanged behavior.
1407
- if (table && table.trim() && seriesIdArg && seriesIdArg.trim()) {
1408
- return genericSeries(table, seriesIdArg, recent, range);
1409
- }
1410
-
1411
- const query = seriesArg?.trim() || seriesIdArg?.trim();
1412
- if (!query) {
1413
- return {
1414
- error: 'not_found',
1415
- message: 'rba_series needs either `series` (a name like "90-day bank bill rate" or an exact series id) or both `table` and `series_id`. Call rba_list_series first if unsure.',
1416
- };
1417
- }
1418
-
1419
- const resolved = await resolveSeriesName(query, table);
1420
- if (!resolved.ok) {
1421
- return { error: resolved.error, message: resolved.message, ...(resolved.candidates ? { candidates: resolved.candidates } : {}) };
1422
- }
1423
- const data = await genericSeries(resolved.table, resolved.series_id, recent, range);
1424
- if ('error' in (data as Record<string, unknown>)) return data; // shouldn't happen post-resolution, but don't mask it
1425
- return { ...(data as object), resolved_from: resolved.resolved_from, resolved_to: resolved.resolved_to };
1426
- }
1427
-
1428
- // ── Router ───────────────────────────────────────────────────────────
1429
-
1430
- async function callTool(name: string, args: Record<string, unknown>): Promise<unknown> {
1431
- switch (name) {
1432
- case 'rba_cash_rate':
1433
- return cashRate(args.recent as number | undefined);
1434
- case 'rba_exchange_rates':
1435
- return exchangeRates(args.currency as string | undefined, args.recent as number | undefined);
1436
- case 'rba_series':
1437
- return rbaSeriesTool(
1438
- args.table as string | undefined,
1439
- args.series_id as string | undefined,
1440
- args.series as string | undefined,
1441
- args.recent as number | undefined,
1442
- args.start_date,
1443
- args.end_date,
1444
- );
1445
- case 'rba_list_series':
1446
- return listSeries(args.table as string | undefined, args.keyword as string | undefined, args.query as string | undefined);
1447
- default:
1448
- throw new Error(`Unknown tool: ${name}`);
1449
- }
1450
- }
1451
-
1452
- export default { tools, callTool, meter: { credits: 1 } } satisfies McpToolExport;