@vocalstack/mcp 1.5.0 → 1.5.2

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/README.md CHANGED
@@ -65,17 +65,17 @@ Both offer the same tools.
65
65
  - `get_transcription`: Read one transcription's status and text, with timestamps and speakers, or its title and summary: as spoken or translated, at once or after waiting for it to finish.
66
66
  - `list_transcriptions`: List this account's transcriptions, newest first, a page at a time: files, links and live ones, including those broadcasts recorded.
67
67
  - `search_transcriptions`: Find what was said across this account's transcriptions, or in one: the lines with given words or a name (mode words), or the passages that answer a question (mode meaning).
68
- - `update_transcription`: Change a transcription: its name, its public share link, its summary, or whether spoken translations use the speakers' own cloned voices.
68
+ - `update_transcription`: Change a transcription: its name, its public share link, its summary, whether spoken translations use the speakers' own cloned voices, or bring it back from the trash.
69
69
  - `delete_transcription`: Move a transcription to the trash, with its recordings, transcript, translations and summary; its share links stop working.
70
70
  - `list_speakers`: List the speakers this account has saved: people recognised by name in every new transcription.
71
- - `save_speaker`: Save one speaker of a transcription to the account's saved speakers, so later transcriptions recognise and name them; their own speech in this transcription is the sample.
72
- - `update_speaker`: Change a speaker: name one person in one transcription (instead of "Speaker N"), or rename a saved speaker and switch their cloned voice on or off, everywhere they are recognised.
71
+ - `update_speaker`: Name or save a speaker: name one person in one transcription (instead of "Speaker N") and optionally save them so later transcriptions recognise them, or rename a saved speaker and switch their cloned voice on or off, everywhere they are recognised.
73
72
  - `delete_speaker`: Delete a saved speaker for good: their name, fingerprint, stored sample and any cloned voice are erased at once and cannot be restored.
74
73
  - `list_broadcasts`: List this account's broadcasts, or get one in detail.
75
74
  - `start_broadcast`: Start live transcription and translation of a stream playing now, on a public link listeners open in their own language.
76
- - `update_broadcast`: Change a broadcast's settings: its name, who can listen (public, or private with a password), the language spoken, and whether translations are read aloud, in stock or the speakers' own cloned voices.
75
+ - `update_broadcast`: Change a broadcast's settings: its name, who can listen (public, or private with a password), the language spoken, and whether translations are read aloud, in stock or the speakers' own cloned voices; or bring it back from the trash.
77
76
  - `stop_broadcast`: Stop the live transcription a broadcast is running, which also stops its billing.
78
77
  - `delete_broadcast`: Move a broadcast to the trash: its public link and stream addresses stop working for listeners.
78
+ - `get_account`: Check the account's credit and usage: the balance left, the plan, whether a new file or live transcription would start now (and what would refuse it), and this period's limits.
79
79
  <!-- /tools -->
80
80
 
81
81
  Errors come back as tool results with `isError` set and a body with `error` (a code such as `not_found`, `invalid_options`, `invalid_state`, `insufficient_credit` or `rate_limited`), `message`, `retryable`, `retryAfterS` and `topUpUrl` when credit is needed. Retry a start with the same `idempotencyKey` so nothing is made or billed twice.
package/dist/handlers.js CHANGED
@@ -701,14 +701,10 @@ async function speakers(_a, ctx) {
701
701
  function noSpeaker(number) {
702
702
  return new ToolError("not_found", `This transcription has no Speaker ${number} who has spoken; get_transcription shows the speaker numbers.`);
703
703
  }
704
- async function saveSpeaker(a, ctx) {
705
- const id = a.id("transcriptionId");
706
- const number = a.int("speaker", 1, 1, 1000);
707
- const name = a.reqStr("name");
708
- if ([...name].length > 60)
709
- throw bad("`name` must be 1 to 60 characters");
710
- const consent = a.bool("consent");
711
- await transcription(ctx, id);
704
+ async function saveSpeaker(ctx, id, number, name, consent) {
705
+ const n = [...name].length;
706
+ if (n === 0 || n > 60)
707
+ throw bad("with `save`, `name` must be 1 to 60 characters");
712
708
  const snap = await snapshot(ctx, id);
713
709
  const spoken = spokenSegments(snap).filter((l) => l.speaker >= 0);
714
710
  const numbers = numbering(spoken, snap);
@@ -719,6 +715,17 @@ async function saveSpeaker(a, ctx) {
719
715
  .sort((x, y) => x - y)[0];
720
716
  if (appearance === undefined)
721
717
  throw noSpeaker(number);
718
+ const known = (await all(ctx, "/v1/speakers", "speakers")).find((s) => s.transcriptionId === id && s.consent !== "refused" && !s.draft && s.library !== false && typeof s.name === "string" && s.name.toLowerCase() === name.toLowerCase());
719
+ if (known) {
720
+ let s = known;
721
+ if (consent && s.consent !== "granted") {
722
+ s = (await ctx.api.request("POST", path("/v1/speakers/", s.id, "/consent"), { json: { consent: "granted" }, signal: ctx.signal })).data;
723
+ }
724
+ const out = { speakerId: s.id, saved: false, consent: s.consent, cloned: s.voice != null };
725
+ if (s.voice)
726
+ put(out, "voiceState", s.voice.state);
727
+ return out;
728
+ }
722
729
  let s;
723
730
  try {
724
731
  s = (await ctx.api.request("POST", "/v1/speakers/from-appearance", {
@@ -731,11 +738,32 @@ async function saveSpeaker(a, ctx) {
731
738
  throw noSpeaker(number);
732
739
  throw e;
733
740
  }
734
- const out = { speakerId: s.id, name: s.name, consent: s.consent, cloned: s.voice != null };
741
+ const out = { speakerId: s.id, saved: true, consent: s.consent, cloned: s.voice != null };
735
742
  if (s.voice)
736
743
  put(out, "voiceState", s.voice.state);
737
744
  return out;
738
745
  }
746
+ async function getAccount(_a, ctx) {
747
+ const b = await ctx.api.get("/v1/billing", {}, ctx.signal);
748
+ const l = b.limitsUsage ?? {};
749
+ const out = {
750
+ billingEnabled: b.enabled ?? null,
751
+ plan: b.plan ?? null,
752
+ balance: b.balance ?? null,
753
+ canStart: b.canStart ?? null,
754
+ startBlock: b.startBlock ?? null,
755
+ canStartLive: b.canStartLive ?? null,
756
+ startBlockLive: b.startBlockLive ?? null,
757
+ running: l.running ?? null,
758
+ concurrentLimit: l.concurrentLimit ?? null,
759
+ createdLastHour: l.createdLastHour ?? null,
760
+ createLimit: l.createLimit ?? null,
761
+ period: b.period ?? null,
762
+ topUpUrl: b.topUpUrl ?? null,
763
+ };
764
+ put(out, "monthlyCredit", b.monthlyCredit);
765
+ return out;
766
+ }
739
767
  async function deleteSpeaker(a, ctx) {
740
768
  const id = a.id("speakerId");
741
769
  try {
@@ -870,6 +898,8 @@ async function patchSpeaker(ctx, id, body) {
870
898
  }
871
899
  async function updateSavedSpeaker(a, ctx) {
872
900
  const id = a.id("speakerId");
901
+ if (present(a, "save"))
902
+ throw bad("`save` is for a speaker in one transcription (transcriptionId with speaker); this one is already saved");
873
903
  let name;
874
904
  const given = a.raw.name;
875
905
  if (given !== undefined && given !== null) {
@@ -908,7 +938,7 @@ async function updateSavedSpeaker(a, ctx) {
908
938
  }
909
939
  const cur = await savedSpeaker(ctx, id);
910
940
  if (cloned === true && cur.consent === "refused") {
911
- throw new ToolError("invalid_state", "This speaker refused to be cloned and their audio was deleted; save them again from a transcription (save_speaker with consent) to clone them.");
941
+ throw new ToolError("invalid_state", "This speaker refused to be cloned and their audio was deleted; save them again from a transcription (update_speaker with save and consent) to clone them.");
912
942
  }
913
943
  if (name !== undefined)
914
944
  await patchSpeaker(ctx, id, { name });
@@ -919,7 +949,7 @@ async function updateSavedSpeaker(a, ctx) {
919
949
  }
920
950
  catch (e) {
921
951
  if (e instanceof ToolError && e.code === "invalid_state" && e.message.includes("no sample stored")) {
922
- throw new ToolError("invalid_state", "This saved speaker has no stored sample to clone their voice from; save them again from a transcription (save_speaker with consent).");
952
+ throw new ToolError("invalid_state", "This saved speaker has no stored sample to clone their voice from; save them again from a transcription (update_speaker with save and consent).");
923
953
  }
924
954
  noSavedSpeaker(e);
925
955
  }
@@ -937,6 +967,11 @@ async function updateSpeaker(a, ctx) {
937
967
  if (present(a, "clonedVoice")) {
938
968
  throw bad("`clonedVoice` is for a saved speaker (speakerId); for everyone in one transcription use update_transcription with clonedVoices");
939
969
  }
970
+ const save = optBool(a, "save") ?? false;
971
+ const consent = a.bool("consent");
972
+ if (consent && !save) {
973
+ throw bad("`consent` with `transcriptionId` goes with `save: true` (saving and cloning this person); to clone a saved speaker use `speakerId` with `clonedVoice`");
974
+ }
940
975
  const id = a.id("transcriptionId");
941
976
  const number = a.int("speaker", 1, 1, 1000);
942
977
  const given = a.raw.name;
@@ -945,7 +980,11 @@ async function updateSpeaker(a, ctx) {
945
980
  if (typeof given !== "string")
946
981
  throw bad('`name` must be a string ("" clears the name)');
947
982
  const name = [...given.trim()].slice(0, 80).join("");
983
+ if (save && (name.length === 0 || [...given.trim()].length > 60))
984
+ throw bad("`name` must be 1 to 60 characters");
948
985
  await transcription(ctx, id);
986
+ if (save)
987
+ return { transcriptionId: id, speaker: number, name, ...(await saveSpeaker(ctx, id, number, given.trim(), consent)) };
949
988
  try {
950
989
  await ctx.api.request("PUT", path("/v1/transcriptions/", id, "/appearances/", String(number)), {
951
990
  json: { name: name || null },
@@ -977,19 +1016,26 @@ async function updateTranscription(a, ctx) {
977
1016
  throw bad("`name` must be 1 to 200 characters");
978
1017
  }
979
1018
  const share = optBool(a, "shareLink");
1019
+ const restore = optBool(a, "restore");
1020
+ if (restore === false)
1021
+ throw bad("`restore` takes only true (bring it back from the trash); delete_transcription moves it there");
980
1022
  const summary = optBool(a, "summary");
981
1023
  const cloned = optBool(a, "clonedVoices");
982
1024
  const consent = a.bool("consent");
983
1025
  if (summary === false)
984
1026
  throw bad("`summary` takes only true (write one now); a summary is never deleted");
985
- if (name === undefined && share === undefined && summary === undefined && cloned === undefined) {
986
- throw bad("give at least one of `name`, `shareLink`, `summary` or `clonedVoices` to change");
1027
+ if (name === undefined && share === undefined && summary === undefined && cloned === undefined && restore === undefined) {
1028
+ throw bad("give at least one of `name`, `shareLink`, `summary`, `clonedVoices` or `restore` to change");
987
1029
  }
988
1030
  if (cloned === true && !consent)
989
1031
  throw bad(NEEDS_CONSENT);
990
1032
  let t = await transcription(ctx, id);
991
- if (t.lifecycle && t.lifecycle !== "active") {
992
- throw new ToolError("invalid_state", "This transcription is in the trash; restore it in the VocalStack Dashboard first.");
1033
+ const restored = restore === true && t.lifecycle === "archived";
1034
+ if (restored) {
1035
+ t = (await ctx.api.request("POST", path("/v1/transcriptions/", id, "/restore"), { signal: ctx.signal })).data;
1036
+ }
1037
+ else if (t.lifecycle && t.lifecycle !== "active") {
1038
+ throw new ToolError("invalid_state", "This transcription is in the trash; pass restore: true to bring it back.");
993
1039
  }
994
1040
  const written = summary === true ? await writeSummary(ctx, id, t) : undefined;
995
1041
  if (name !== undefined || cloned !== undefined) {
@@ -1016,6 +1062,8 @@ async function updateTranscription(a, ctx) {
1016
1062
  links = [];
1017
1063
  }
1018
1064
  const out = { transcriptionId: t.id, name: t.name, shareUrl: links[0]?.url ?? null, clonedVoices: t.settings?.voiceCloning === true };
1065
+ if (restore !== undefined)
1066
+ out.restored = restored;
1019
1067
  if (written)
1020
1068
  out.summary = written;
1021
1069
  return out;
@@ -1029,12 +1077,19 @@ async function updateBroadcast(a, ctx) {
1029
1077
  const speak = optBool(a, "spokenTranslations");
1030
1078
  const cloned = optBool(a, "clonedVoices");
1031
1079
  const consent = a.bool("consent");
1032
- if ([name, access, password, spokenRaw, speak, cloned].every((v) => v === undefined)) {
1033
- throw bad("give at least one of `name`, `access`, `password`, `spokenLanguage`, `spokenTranslations` or `clonedVoices` to change");
1080
+ const restore = optBool(a, "restore");
1081
+ if (restore === false)
1082
+ throw bad("`restore` takes only true (bring it back from the trash); delete_broadcast moves it there");
1083
+ const changes = [name, access, password, spokenRaw, speak, cloned].some((v) => v !== undefined);
1084
+ if (!changes && restore === undefined) {
1085
+ throw bad("give at least one of `name`, `access`, `password`, `spokenLanguage`, `spokenTranslations`, `clonedVoices` or `restore` to change");
1034
1086
  }
1035
1087
  if (cloned === true && !consent)
1036
1088
  throw bad(NEEDS_CONSENT);
1037
- await broadcast(ctx, id);
1089
+ let b = await broadcast(ctx, id);
1090
+ const restored = restore === true && b.lifecycle === "archived";
1091
+ if (restored)
1092
+ b = (await ctx.api.request("POST", path("/v1/broadcasts/", id, "/restore"), { signal: ctx.signal })).data;
1038
1093
  const body = {};
1039
1094
  if (name !== undefined)
1040
1095
  body.name = name;
@@ -1055,24 +1110,25 @@ async function updateBroadcast(a, ctx) {
1055
1110
  }
1056
1111
  if (Object.keys(settings).length)
1057
1112
  body.settings = settings;
1058
- let b;
1059
- try {
1060
- b = (await ctx.api.request("PATCH", path("/v1/broadcasts/", id), { json: body, signal: ctx.signal })).data;
1061
- }
1062
- catch (e) {
1063
- if (e instanceof ToolError && e.code === "invalid_state") {
1064
- throw new ToolError("invalid_state", "This broadcast is in the trash; restore it in the VocalStack Dashboard first.");
1113
+ if (changes)
1114
+ try {
1115
+ b = (await ctx.api.request("PATCH", path("/v1/broadcasts/", id), { json: body, signal: ctx.signal })).data;
1065
1116
  }
1066
- if (e instanceof ToolError && e.message === "private broadcasts need a password") {
1067
- throw bad("A private broadcast needs a `password` (at least 4 characters) the first time.");
1117
+ catch (e) {
1118
+ if (e instanceof ToolError && e.code === "invalid_state") {
1119
+ throw new ToolError("invalid_state", "This broadcast is in the trash; pass restore: true to bring it back.");
1120
+ }
1121
+ if (e instanceof ToolError && e.message === "private broadcasts need a password") {
1122
+ throw bad("A private broadcast needs a `password` (at least 4 characters) the first time.");
1123
+ }
1124
+ consentError(e);
1068
1125
  }
1069
- consentError(e);
1070
- }
1071
1126
  return {
1072
1127
  ...brief(b),
1073
1128
  spokenLanguage: b.settings?.spokenLanguage ?? "auto",
1074
1129
  spokenTranslations: b.settings?.translationTts === true,
1075
1130
  clonedVoices: b.settings?.voiceCloning === true,
1131
+ ...(restore !== undefined ? { restored } : {}),
1076
1132
  };
1077
1133
  }
1078
1134
  const ON_AIR = "This broadcast is already on air; stop it first (stop_broadcast) or use another broadcast.";
@@ -1191,7 +1247,6 @@ export const handlers = {
1191
1247
  update_transcription: updateTranscription,
1192
1248
  delete_transcription: deleteTranscription,
1193
1249
  list_speakers: speakers,
1194
- save_speaker: saveSpeaker,
1195
1250
  update_speaker: updateSpeaker,
1196
1251
  delete_speaker: deleteSpeaker,
1197
1252
  list_broadcasts: listBroadcasts,
@@ -1199,4 +1254,5 @@ export const handlers = {
1199
1254
  update_broadcast: updateBroadcast,
1200
1255
  stop_broadcast: stopBroadcast,
1201
1256
  delete_broadcast: deleteBroadcast,
1257
+ get_account: getAccount,
1202
1258
  };
package/dist/tools.json CHANGED
@@ -1,10 +1,10 @@
1
1
  {
2
- "instructions": "VocalStack transcribes and translates recorded or live audio and video in about 100 languages, with speaker labels.\nRecorded audio: create_transcription (a link or a file) returns a transcriptionId at once; get_transcription with waitS waits for it and reads the text, in any language with lang.\nFind: list_transcriptions (ids, newest first); search_transcriptions (mode words for words or names, meaning for a question).\nOne transcription: get_transcription with format summary for its title and summary; update_transcription (name, share link, summary, cloned voices); delete_transcription (to the trash, restorable in the Dashboard).\nLive streams: start_broadcast, then get_transcription while it runs, stop_broadcast to end it; list_broadcasts shows them; update_broadcast changes access, language and spoken translations; delete_broadcast removes one.\nPeople: list_speakers, save_speaker (voice cloned only with the person's consent), update_speaker (names, and a saved speaker's cloned voice), delete_speaker.\nOnly transcription is billed; everything else, and listening, is free. Retry a start with the same idempotencyKey.\nErrors are results with isError true: error, message, retryable, and retryAfterS or topUpUrl when they apply.",
2
+ "instructions": "VocalStack transcribes and translates recorded or live audio and video in about 100 languages, with speaker labels.\nRecorded audio: create_transcription (a link or a file) returns a transcriptionId at once; get_transcription with waitS waits for it and reads the text, in any language with lang.\nFind: list_transcriptions (ids, newest first); search_transcriptions (mode words for words or names, meaning for a question).\nOne transcription: get_transcription with format summary for its title and summary; update_transcription (name, share link, summary, cloned voices); delete_transcription (to the trash; update_transcription with restore brings it back).\nLive streams: start_broadcast, then get_transcription while it runs, stop_broadcast to end it; list_broadcasts shows them; update_broadcast changes access, language and spoken translations; delete_broadcast removes one.\nPeople: list_speakers, update_speaker (names; save true remembers a person, their voice cloned only with their consent; a saved speaker's cloned voice), delete_speaker.\nCredit and usage: get_account (balance, plan, whether a new transcription would start, limits).\nOnly transcription is billed; everything else, and listening, is free. Retry a start with the same idempotencyKey.\nErrors are results with isError true: error, message, retryable, and retryAfterS or topUpUrl when they apply.",
3
3
  "tools": [
4
4
  {
5
5
  "name": "create_transcription",
6
6
  "title": "Create a transcription",
7
- "description": "Start transcribing a finished recording, from a public link or a file sent inline; returns a transcriptionId at once, before any text.\nUse: audio or video that has ended (a file, podcast episode, YouTube video or public post). A stream playing now: start_broadcast. The result: get_transcription with waitS, no polling needed.\nList price: $0.20 an hour of audio, billed per second transcribed; Pay as you go bills at least 5 minutes per transcription. Each call is a new, billed transcription unless it repeats an idempotencyKey; a refused call costs nothing; summary and shareLink are free.\nNew transcriptions an hour per account: 10 on Pay as you go, 100 on Premium.\nTranscriptions running at once per account: 5 on Pay as you go, 50 on Premium; more wait their turn instead of failing.\nReturns: transcriptionId, state (queued, fetching, transcribing, then done, failed or stopped), progress, shareUrl when shareLink is true, replayed true for a repeated key.\nErrors: invalid_options (neither or both of url and fileBase64, not a public http(s) link, bad base64, too large), invalid_audio (empty file), idempotency_key_reused, insufficient_credit (with topUpUrl), quota_exceeded, rate_limited (with retryAfterS). A source that cannot be read ends in state failed, with error.",
7
+ "description": "Start transcribing a finished recording, from a public link or a file sent inline; returns a transcriptionId at once, before any text.\nUse: audio or video that has ended (a file, podcast episode, YouTube video or public post). A stream playing now: start_broadcast. The result: get_transcription with waitS, no polling needed.\nArguments: exactly one source: url, or fileBase64 with filename (over 10 MiB: host the file and give url). startS and endS clip both what is transcribed and what is billed. spokenLanguage helps short or noisy audio; speakerLabels is fixed once created, while summary and shareLink can also be turned on later with update_transcription.\nBehaviour: VocalStack's servers download the link once, as a visitor who is not signed in; nothing is posted to or changed on the source site.\nList price: $0.20 an hour of audio, billed per second transcribed; Pay as you go bills at least 5 minutes per transcription. Each call is a new, billed transcription unless it repeats an idempotencyKey; a refused call costs nothing; summary and shareLink are free.\nNew transcriptions an hour per account: 10 on Pay as you go, 100 on Premium.\nTranscriptions running at once per account: 5 on Pay as you go, 50 on Premium; more wait their turn instead of failing.\nReturns: transcriptionId, state (queued, fetching, transcribing, then done, failed or stopped), progress, shareUrl when shareLink is true, replayed true for a repeated key.\nErrors: invalid_options (neither or both of url and fileBase64, not a public http(s) link, bad base64, too large), invalid_audio (empty file), idempotency_key_reused, insufficient_credit (with topUpUrl), quota_exceeded, rate_limited (with retryAfterS). A source that cannot be read ends in state failed, with error.",
8
8
  "inputSchema": {
9
9
  "additionalProperties": false,
10
10
  "properties": {
@@ -111,7 +111,7 @@
111
111
  {
112
112
  "name": "get_transcription",
113
113
  "title": "Read a transcription",
114
- "description": "Read one transcription's status and text, with timestamps and speakers, or its title and summary: as spoken or translated, at once or after waiting for it to finish.\nUse: after create_transcription (waitS saves polling), or any time during a live transcription for the text so far. An overview: format summary. Finding a passage in many transcriptions: search_transcriptions.\nBehaviour: free. A wait ends as soon as the work does, and sends progress notifications when the client gives a progress token. lang needs a finished transcription; the first call in a new language can take up to a minute on a long one, later pages are quick. Page with offset = nextOffset.\nReturns: status (state, progress, error when failed, shareUrl when it has a public link: made with shareLink in create_transcription or update_transcription), timedOut when waitS was set, then segmentCount, partial while text is still coming, text or segments, nextOffset when more remain, and lang and translated with lang. format summary returns summary instead: state ready with title and summary (and lang when translated), pending while it is written or the transcription runs, none when none was asked for (update_transcription with summary true writes one), not_requested under a minute, unavailable when it failed. A wait that runs out with lang set returns the status only.\nErrors: not_found; invalid_state for lang before it is finished; invalid_options for an unsupported lang or includeWords with lang; rate_limited when waitS is set and 4 waits are already open for this key (8 for the account).",
114
+ "description": "Read one transcription's status and text, with timestamps and speakers, or its title and summary: as spoken or translated, at once or after waiting for it to finish.\nUse: after create_transcription (waitS saves polling), or any time during a live transcription for the text so far. An overview: format summary. Finding a passage in many transcriptions: search_transcriptions.\nBehaviour: free. A wait ends as soon as the work does, and sends progress notifications when the client gives a progress token. lang needs a finished transcription; the first call in a new language can take up to a minute on a long one, later pages are quick. Page with offset = nextOffset.\nReturns: status (state, progress, error when failed, shareUrl when it has a public link: made with shareLink in create_transcription or update_transcription), timedOut when waitS was set, then segmentCount, partial while text is still coming, text or segments, nextOffset when more remain, and lang and translated with lang. format summary returns summary instead: state ready with title and summary (and lang when translated), pending while it is written or the transcription runs, none when none was asked for (update_transcription with summary true writes one), not_requested under a minute, unavailable when it failed. A wait that runs out with lang set returns the status only.\nErrors: not_found; invalid_state for lang before it is finished; invalid_options for an unsupported lang or includeWords with lang; rate_limited (with retryAfterS) when waitS is set and 4 waits are already open for this key (8 for the account), or past 600 calls a minute per API key.",
115
115
  "inputSchema": {
116
116
  "additionalProperties": false,
117
117
  "properties": {
@@ -192,7 +192,7 @@
192
192
  {
193
193
  "name": "list_transcriptions",
194
194
  "title": "List transcriptions",
195
- "description": "List this account's transcriptions, newest first, a page at a time: files, links and live ones, including those broadcasts recorded.\nUse: to find a transcriptionId or see what exists. What was said: search_transcriptions. Broadcasts: list_broadcasts.\nBehaviour: free. lifecycle archived lists the trash: deleted transcriptions, which the VocalStack Dashboard can restore for 30 days. A cursor this tool did not give out is refused.\nReturns: transcriptions (transcriptionId, name, createdAt, kind file or live, state, durationS and title when known, broadcastId for a broadcast's), and nextCursor while more remain. None: an empty list.\nErrors: invalid_options for an unknown cursor.",
195
+ "description": "List this account's transcriptions, newest first, a page at a time: files, links and live ones, including those broadcasts recorded.\nUse: to find a transcriptionId or see what exists. What was said: search_transcriptions. Broadcasts: list_broadcasts.\nBehaviour: free. Ordered by creation time, so pages are stable: transcriptions made or deleted while paging never repeat or shift another out of its page. lifecycle archived lists the trash: deleted transcriptions, restorable for 30 days with update_transcription restore true. A cursor this tool did not give out is refused.\nReturns: transcriptions (transcriptionId, name, createdAt, kind file or live, state, durationS and title when known, broadcastId for a broadcast's), and nextCursor while more remain. None: an empty list.\nErrors: invalid_options for an unknown cursor; rate_limited (with retryAfterS) past 600 calls a minute per API key.",
196
196
  "inputSchema": {
197
197
  "additionalProperties": false,
198
198
  "properties": {
@@ -233,7 +233,7 @@
233
233
  {
234
234
  "name": "search_transcriptions",
235
235
  "title": "Search transcriptions",
236
- "description": "Find what was said across this account's transcriptions, or in one: the lines with given words or a name (mode words), or the passages that answer a question (mode meaning).\nUse: words when you know what was said; meaning for a question or topic, in any language. Text around a hit: get_transcription with offset. One transcription's overview: get_transcription with format summary.\nBehaviour: free; no paging. words ranks whole words first, then close spellings, then newest, and searches the query's English translation too. meaning returns up to 10 passages, closest first. Searches by meaning an hour per account: 60 on Pay as you go, no limit on Premium.\nReturns: words: query, queryEn, hits (transcriptionId, name, mediaStart in seconds, speaker, text, score, live, recordedAt, rendered with lang). meaning: query, state (ready, or unavailable or off when it cannot run), passages (transcriptionId, name, mediaStart, mediaEnd, text, similarity 0 to 1, live), searchesLeftThisHour when capped. Cite the transcription and second. No match: an empty list.\nErrors: not_found for an unknown transcriptionId; invalid_options for lang or limit with mode meaning; rate_limited (with retryAfterS) past the hourly searches by meaning.",
236
+ "description": "Find what was said across this account's transcriptions, or in one: the lines with given words or a name (mode words), or the passages that answer a question (mode meaning).\nUse: words when you know what was said; meaning for a question or topic, in any language. Text around a hit: get_transcription with offset. One transcription's overview: get_transcription with format summary.\nArguments: query is phrased differently per mode: a few words or a name for words, a whole question for meaning. transcriptionId narrows either mode to one transcription. lang and limit work with words only, and lang changes only how hits are shown, never which lines match.\nBehaviour: free; no paging. words ranks whole words first, then close spellings, then newest, and searches the query's English translation too, and finds a line once it is final, so the last seconds of a live transcription may not show yet. The trash is never searched. meaning returns up to 10 passages, closest first, and covers finished recordings only (they are indexed after they end). Searches by meaning an hour per account: 60 on Pay as you go, no limit on Premium.\nReturns: words: query, queryEn, hits (transcriptionId, name, mediaStart in seconds, speaker, text, score, live, recordedAt, rendered with lang). meaning: query, state (ready, or unavailable or off when it cannot run), passages (transcriptionId, name, mediaStart, mediaEnd, text, similarity 0 to 1, live), searchesLeftThisHour when capped. Cite the transcription and second. No match: an empty list.\nErrors: not_found for an unknown transcriptionId; invalid_options for lang or limit with mode meaning; rate_limited (with retryAfterS) past the hourly searches by meaning.",
237
237
  "inputSchema": {
238
238
  "additionalProperties": false,
239
239
  "properties": {
@@ -298,7 +298,7 @@
298
298
  {
299
299
  "name": "update_transcription",
300
300
  "title": "Update a transcription",
301
- "description": "Change a transcription: its name, its public share link, its summary, or whether spoken translations use the speakers' own cloned voices.\nUse: renaming, sharing or unsharing, asking for a summary later, or switching cloned voices. Only the fields given change; give at least one. Speaker names: update_speaker. Removing it: delete_transcription.\nArguments: transcriptionId (required). name 1 to 200 characters. shareLink true makes a public read-only link (or keeps the one it has), false turns every public link off. summary true writes a title and summary if it has none (a finished transcription of at least a minute). clonedVoices true needs consent true, recording that everyone speaking agreed to be cloned; false needs no consent.\nBehaviour: free; calling again with the same fields changes nothing. A link turned off cannot be turned back on: shareLink true afterwards makes a new address. With cloned voices on, each speaker's translated words are read in that speaker's own voice, never anyone else's; speakers who refused, or without a usable clone, keep a stock voice; lines read after the change use it.\nReturns: transcriptionId, name, shareUrl (null with no public link), summary (state ready, pending, not_requested or unavailable) when asked for, clonedVoices.\nErrors: invalid_options for no field to change, a bad name, summary false, or clonedVoices true without consent; invalid_state for a deleted transcription, or summary before it is finished; not_found; forbidden for a read-only key.",
301
+ "description": "Change a transcription: its name, its public share link, its summary, whether spoken translations use the speakers' own cloned voices, or bring it back from the trash.\nUse: renaming, sharing or unsharing, asking for a summary later, switching cloned voices, or restoring one deleted by mistake. Only the fields given change; give at least one. Speaker names: update_speaker. Removing it: delete_transcription.\nArguments: transcriptionId (required). name 1 to 200 characters. shareLink true makes a public read-only link (or keeps the one it has), false turns every public link off. summary true writes a title and summary if it has none (a finished transcription of at least a minute). clonedVoices true needs consent true, recording that everyone speaking agreed to be cloned; false needs no consent. restore true brings a deleted one back first, then applies the rest.\nBehaviour: free; calling again with the same fields changes nothing. A link turned off cannot be turned back on: shareLink true afterwards makes a new address. With cloned voices on, each speaker's translated words are read in that speaker's own voice, never anyone else's; speakers who refused, or without a usable clone, keep a stock voice; lines read after the change use it.\nReturns: transcriptionId, name, shareUrl (null with no public link), summary (state ready, pending, not_requested or unavailable) when asked for, clonedVoices, restored with restore.\nErrors: invalid_options for no field to change, a bad name, summary or restore false, or clonedVoices true without consent; invalid_state for a deleted transcription without restore true, or summary before it is finished; not_found once erased after the restore window; not_found; forbidden for a read-only key.",
302
302
  "inputSchema": {
303
303
  "additionalProperties": false,
304
304
  "properties": {
@@ -320,6 +320,13 @@
320
320
  "minLength": 1,
321
321
  "type": "string"
322
322
  },
323
+ "restore": {
324
+ "description": "true: bring it back from the trash (after delete_transcription, within the restore window) with everything it had; other fields given apply after. false is not accepted.",
325
+ "examples": [
326
+ true
327
+ ],
328
+ "type": "boolean"
329
+ },
323
330
  "shareLink": {
324
331
  "description": "true: make a public link anyone can open to read the transcript without signing in (an existing one is kept; shareUrl in the result). false: every public link of it stops working at once.",
325
332
  "type": "boolean"
@@ -353,7 +360,7 @@
353
360
  {
354
361
  "name": "delete_transcription",
355
362
  "title": "Delete a transcription",
356
- "description": "Move a transcription to the trash, with its recordings, transcript, translations and summary; its share links stop working.\nUse: to remove one. For 30 days it can be restored unchanged from the trash in the VocalStack Dashboard (list_transcriptions with lifecycle archived lists it); after that it is erased for good. Saved speakers stay (delete_speaker). A broadcast itself: delete_broadcast.\nBehaviour: a file or link still being transcribed is stopped first and keeps its text so far, billed only for what was transcribed; restoring does not resume it. A live one must be stopped first (stop_broadcast).\nReturns: transcriptionId, name, deleted, deletedAt, restorableUntil.\nErrors: invalid_state while recording live or when locked; not_found; forbidden for a read-only key.",
363
+ "description": "Move a transcription to the trash, with its recordings, transcript, translations and summary; its share links stop working.\nUse: to remove one. For 30 days update_transcription with restore true brings it back unchanged (list_transcriptions with lifecycle archived lists the trash); after that it is erased for good. Saved speakers stay (delete_speaker). A broadcast itself: delete_broadcast.\nArguments: transcriptionId, a 36-character UUID (8-4-4-4-12 hex digits, as 0190f0e0-7c1a-7d3e-9b2a-5f4e3d2c1b0a) from create_transcription, list_transcriptions or search_transcriptions; a broadcastId, speakerId or the code in a share link is not_found. One already in the trash is left as it is, so a retry is safe.\nBehaviour: free, and a soft delete: nothing is erased for 30 days. At once its share links answer not found, voices reading its translations stop, and it drops out of list_transcriptions and search_transcriptions. Restoring brings back the same share links. A file or link still being transcribed is stopped first and keeps its text so far, billed only for what was transcribed; restoring does not resume it. A live one must be stopped first (stop_broadcast).\nReturns: transcriptionId, name, deleted, deletedAt, restorableUntil.\nErrors: invalid_state while recording live or when locked; not_found; forbidden for a read-only key.",
357
364
  "inputSchema": {
358
365
  "additionalProperties": false,
359
366
  "properties": {
@@ -382,7 +389,7 @@
382
389
  {
383
390
  "name": "list_speakers",
384
391
  "title": "List saved speakers",
385
- "description": "List the speakers this account has saved: people recognised by name in every new transcription.\nUse: to see who is known or find a speakerId for delete_speaker. Save one: save_speaker. Names inside one transcription come from get_transcription and are set with update_speaker, which saves no one.\nBehaviour: free; no paging.\nReturns: speakers, newest first, each with id, name, gender when known, cloned (their voice was cloned, with consent, to read translations of their own words) and, when cloned, voiceReady. None saved: an empty list.",
392
+ "description": "List the speakers this account has saved: people recognised by name in every new transcription.\nUse: to see who is known, before saving someone (update_speaker with save), or to find a speakerId for update_speaker or delete_speaker. Names inside one transcription come from get_transcription.\nBehaviour: free; no paging.\nReturns: speakers, newest first, each with id, name, gender when known, cloned (their voice was cloned, with consent, to read translations of their own words) and, when cloned, voiceReady. None saved: an empty list.",
386
393
  "inputSchema": {
387
394
  "additionalProperties": false,
388
395
  "properties": {},
@@ -396,63 +403,10 @@
396
403
  "openWorldHint": false
397
404
  }
398
405
  },
399
- {
400
- "name": "save_speaker",
401
- "title": "Save a speaker",
402
- "description": "Save one speaker of a transcription to the account's saved speakers, so later transcriptions recognise and name them; their own speech in this transcription is the sample.\nUse: remembering a person across transcriptions. This transcription only: update_speaker. Who is saved: list_speakers. Remove: delete_speaker.\nArguments: transcriptionId and speaker point at one person: speaker is the number in \"Speaker 2\" from get_transcription, who must have spoken (a live transcription can still be running). name is 1 to 60 characters, trimmed; renaming later: update_speaker with speakerId. consent is true only when that person agreed to be cloned; leaving it out or false keeps only their fingerprint.\nBehaviour: free; also names them in this transcription. Each call saves a new speaker, so check list_speakers first. Without consent only a voice fingerprint is kept; with it the voice is cloned too, recorded with your account and the time.\nReturns: speakerId, name, consent (granted or pending), cloned, voiceState (training, then ready) when cloned.\nErrors: not_found for an unknown transcription or speaker number; invalid_options for too little speech of them (the message says how much there is); forbidden for a read-only key.",
403
- "inputSchema": {
404
- "additionalProperties": false,
405
- "properties": {
406
- "consent": {
407
- "default": false,
408
- "description": "true only when this person agreed to have their voice cloned: it is then cloned too, from their own speech in this transcription, so translations of their words can be read in their voice. false: only their fingerprint is saved. Never set it without the person's agreement.",
409
- "type": "boolean"
410
- },
411
- "name": {
412
- "description": "The person's name, shown in this and later transcriptions.",
413
- "examples": [
414
- "Maria Lopez"
415
- ],
416
- "maxLength": 60,
417
- "minLength": 1,
418
- "type": "string"
419
- },
420
- "speaker": {
421
- "description": "The N of \"Speaker N\" in get_transcription's text, or speakerNumber in its segments. Works while a live transcription is still running.",
422
- "examples": [
423
- 2
424
- ],
425
- "minimum": 1,
426
- "type": "integer"
427
- },
428
- "transcriptionId": {
429
- "description": "A transcriptionId from create_transcription, list_transcriptions, search_transcriptions or start_broadcast.",
430
- "examples": [
431
- "0190f0e0-7c1a-7d3e-9b2a-5f4e3d2c1b0a"
432
- ],
433
- "format": "uuid",
434
- "type": "string"
435
- }
436
- },
437
- "required": [
438
- "transcriptionId",
439
- "speaker",
440
- "name"
441
- ],
442
- "type": "object"
443
- },
444
- "annotations": {
445
- "title": "Save a speaker",
446
- "readOnlyHint": false,
447
- "destructiveHint": false,
448
- "idempotentHint": false,
449
- "openWorldHint": false
450
- }
451
- },
452
406
  {
453
407
  "name": "update_speaker",
454
- "title": "Update a speaker",
455
- "description": "Change a speaker: name one person in one transcription (instead of \"Speaker N\"), or rename a saved speaker and switch their cloned voice on or off, everywhere they are recognised.\nUse: labelling who spoke here, correcting a saved speaker's name, or turning their cloned voice on with their consent. To recognise a person in later transcriptions: save_speaker (it also names them here). Saved speakers: list_speakers.\nArguments: give exactly one of transcriptionId with speaker (the number in \"Speaker 2\" from get_transcription, not a name or an id), with name; or speakerId (from list_speakers) with name, clonedVoice or both. A name in a transcription is cut to 80 characters and an empty one undoes it; a saved speaker's is 1 to 60. clonedVoice true needs consent true.\nBehaviour: free and reversible; calling again replaces the name. In a transcription it applies to that person in every recording of it, and listeners following live see it at once; their speaker number never changes. clonedVoice true clones a saved speaker from their stored sample if they have no clone yet (voiceState training, then ready); false keeps the clone, unused (delete_speaker erases it). Nothing else about the person changes.\nReturns: transcriptionId, speaker, name (null when cleared); or, with speakerId, speakerId, name, consent, clonedVoice, and voiceState when cloned.\nErrors: invalid_options for both or neither kind of target, nothing to change, a bad name, clonedVoice with transcriptionId, or clonedVoice true without consent; invalid_state for a saved speaker who refused cloning or has no sample; not_found for an unknown transcription, speaker number or speakerId; forbidden for a read-only key.",
408
+ "title": "Name or save a speaker",
409
+ "description": "Name or save a speaker: name one person in one transcription (instead of \"Speaker N\") and optionally save them so later transcriptions recognise them, or rename a saved speaker and switch their cloned voice on or off, everywhere they are recognised.\nUse: labelling who spoke here; remembering a person across transcriptions (save true); correcting a saved speaker's name or turning their cloned voice on with their consent. Who is saved: list_speakers. Forgetting someone: delete_speaker.\nArguments: give exactly one target: transcriptionId with speaker (the number in \"Speaker 2\" from get_transcription, not a name or an id) and name, plus save and consent; or speakerId (from list_speakers) with name, clonedVoice or both. A name in a transcription is cut to 80 characters and an empty one undoes it; a saved speaker's, or one with save, is 1 to 60. clonedVoice true needs consent true. Examples: {transcriptionId, speaker: 2, name: \"Maria Lopez\"}; {transcriptionId, speaker: 2, name: \"Maria Lopez\", save: true}; {speakerId, clonedVoice: true, consent: true}.\nBehaviour: free; calling again replaces the name and saves no one twice. In a transcription it applies to that person in every recording of it, and listeners following live see it at once; their speaker number never changes. save keeps their voice fingerprint (and, with consent, clones their voice from their speech here, voiceState training then ready); the person must have spoken, and a live transcription can still be running. clonedVoice true clones a saved speaker from their stored sample if they have no clone yet; false keeps the clone, unused (delete_speaker erases it). Nothing else about the person changes.\nReturns: transcriptionId, speaker, name (null when cleared), and with save speakerId, saved (false when they were already saved), consent, cloned and voiceState; or, with speakerId, speakerId, name, consent, clonedVoice, and voiceState when cloned.\nErrors: invalid_options for both or neither kind of target, nothing to change, a bad name, save or clonedVoice with the wrong target, too little of their speech to save (the message says how much there is), or clonedVoice true without consent; invalid_state for a saved speaker who refused cloning or has no sample; not_found for an unknown transcription, speaker number or speakerId; forbidden for a read-only key.",
456
410
  "inputSchema": {
457
411
  "additionalProperties": false,
458
412
  "properties": {
@@ -462,11 +416,11 @@
462
416
  },
463
417
  "consent": {
464
418
  "default": false,
465
- "description": "Must be true with clonedVoices true: confirms this person agreed to have their voice cloned. Recorded with your account and the time. Not needed to turn cloned voices off.",
419
+ "description": "true only when this person agreed to have their voice cloned: needed with clonedVoice true, and with save it clones them too, from their own speech in this transcription. Recorded with your account and the time. Never set it without the person's agreement; not needed to turn a cloned voice off.",
466
420
  "type": "boolean"
467
421
  },
468
422
  "name": {
469
- "description": "The name to show instead of \"Speaker N\", trimmed: up to 80 characters in one transcription (an empty string clears it back to \"Speaker N\"), 1 to 60 for a saved speaker. Required with transcriptionId.",
423
+ "description": "The name to show instead of \"Speaker N\", trimmed: up to 80 characters in one transcription (an empty string clears it back to \"Speaker N\"), 1 to 60 for a saved speaker or with save. Required with transcriptionId.",
470
424
  "examples": [
471
425
  "Maria Lopez",
472
426
  ""
@@ -474,6 +428,11 @@
474
428
  "maxLength": 80,
475
429
  "type": "string"
476
430
  },
431
+ "save": {
432
+ "default": false,
433
+ "description": "With transcriptionId and speaker only: true also saves this person to the account's saved speakers, so later transcriptions recognise and name them; their own speech here is the sample. A person already saved from this transcription is not saved twice.",
434
+ "type": "boolean"
435
+ },
477
436
  "speaker": {
478
437
  "description": "With transcriptionId: the N of \"Speaker N\" in get_transcription's text, or speakerNumber in its segments; it stays the same after naming.",
479
438
  "examples": [
@@ -483,7 +442,7 @@
483
442
  "type": "integer"
484
443
  },
485
444
  "speakerId": {
486
- "description": "Instead of transcriptionId and speaker: a saved speaker's id (id in list_speakers, or speakerId from save_speaker), changed everywhere they are recognised.",
445
+ "description": "Instead of transcriptionId and speaker: a saved speaker's id (id in list_speakers, or speakerId from update_speaker with save), changed everywhere they are recognised.",
487
446
  "examples": [
488
447
  "0190f0e0-7c1a-7d3e-9b2a-5f4e3d2c1b0a"
489
448
  ],
@@ -502,7 +461,7 @@
502
461
  "type": "object"
503
462
  },
504
463
  "annotations": {
505
- "title": "Update a speaker",
464
+ "title": "Name or save a speaker",
506
465
  "readOnlyHint": false,
507
466
  "destructiveHint": false,
508
467
  "idempotentHint": true,
@@ -512,12 +471,12 @@
512
471
  {
513
472
  "name": "delete_speaker",
514
473
  "title": "Delete a saved speaker",
515
- "description": "Delete a saved speaker for good: their name, fingerprint, stored sample and any cloned voice are erased at once and cannot be restored.\nUse: the person should no longer be recognised, or asked for their voice to be removed. A name inside one transcription: update_speaker.\nArguments: speakerId is a saved speaker's id, the id field of list_speakers or the speakerId save_speaker returned; not a speaker number (\"Speaker 2\") and not a transcriptionId. To unname someone in one transcription instead: update_speaker with an empty name.\nBehaviour: free. Transcripts keep every line they spoke; lines labelled with their saved name go back to \"Speaker N\".\nReturns: speakerId, deleted true.\nErrors: not_found for an unknown or already deleted id; forbidden for a read-only key.",
474
+ "description": "Delete a saved speaker for good: their name, fingerprint, stored sample and any cloned voice are erased at once and cannot be restored.\nUse: the person should no longer be recognised, or asked for their voice to be removed. A name inside one transcription: update_speaker.\nArguments: speakerId is a saved speaker's id, the id field of list_speakers or the speakerId update_speaker with save returned; not a speaker number (\"Speaker 2\") and not a transcriptionId. To unname someone in one transcription instead: update_speaker with an empty name.\nBehaviour: free. Transcripts keep every line they spoke; lines labelled with their saved name go back to \"Speaker N\".\nReturns: speakerId, deleted true.\nErrors: not_found for an unknown or already deleted id; forbidden for a read-only key.",
516
475
  "inputSchema": {
517
476
  "additionalProperties": false,
518
477
  "properties": {
519
478
  "speakerId": {
520
- "description": "A saved speaker's id: id in list_speakers, or speakerId from save_speaker.",
479
+ "description": "A saved speaker's id: id in list_speakers, or speakerId from update_speaker with save.",
521
480
  "examples": [
522
481
  "0190f0e0-7c1a-7d3e-9b2a-5f4e3d2c1b0a"
523
482
  ],
@@ -541,7 +500,7 @@
541
500
  {
542
501
  "name": "list_broadcasts",
543
502
  "title": "List broadcasts",
544
- "description": "List this account's broadcasts, or get one in detail. A broadcast is a public link where listeners follow a live transcription in their own language.\nUse: to find a broadcastId for start_broadcast, stop_broadcast or delete_broadcast, see which are on air, or (with broadcastId) a broadcast's recorded transcriptions and streams. Text: get_transcription; they also appear in list_transcriptions.\nBehaviour: free; most recently updated first, up to 500 per lifecycle, no paging.\nReturns: broadcasts (broadcastId, name, link, access, onAir, listenersNow, lastRecordedAt once it has recorded, archived for one in the trash; while on air transcriptionId, state, onAirSince). With broadcastId, one item that adds transcriptions (up to 20: transcriptionId, name, recording, lastRecordedAt) and streams (name, url). None: an empty list.\nErrors: not_found for an unknown broadcastId.",
503
+ "description": "List this account's broadcasts, or get one in detail. A broadcast is a public link where listeners follow a live transcription in their own language.\nUse: to find a broadcastId for start_broadcast, stop_broadcast or delete_broadcast, see which are on air, or (with broadcastId) a broadcast's recorded transcriptions and streams. Text: get_transcription; they also appear in list_transcriptions.\nBehaviour: free; most recently updated first, up to 500 per lifecycle, no paging; rate_limited (with retryAfterS) past 600 calls a minute per API key.\nReturns: broadcasts (broadcastId, name, link, access, onAir, listenersNow, lastRecordedAt once it has recorded, archived for one in the trash; while on air transcriptionId, state, onAirSince). With broadcastId, one item that adds transcriptions (up to 20: transcriptionId, name, recording, lastRecordedAt) and streams (name, url). None: an empty list.\nErrors: not_found for an unknown broadcastId.",
545
504
  "inputSchema": {
546
505
  "additionalProperties": false,
547
506
  "properties": {
@@ -576,7 +535,7 @@
576
535
  {
577
536
  "name": "start_broadcast",
578
537
  "title": "Start a broadcast",
579
- "description": "Start live transcription and translation of a stream playing now, on a public link listeners open in their own language.\nUse: a stream playing now. A recording that has ended: create_transcription. Reading it while it runs: get_transcription.\nArguments: streamUrl must be live when called; the same streamUrl on the same broadcast reuses its stream. broadcastId (from list_broadcasts) reruns an existing broadcast on its old link, and then name is ignored; without it a new broadcast is made, named name or the stream's host. spokenLanguage is a code like en (omit to detect). spokenTranslations true adds audio of each translation. idempotencyKey: any string you choose, reused only to retry this same call.\nList price: $0.60 an hour while it runs, billed per second; Pay as you go bills at least 5 minutes per transcription. Listeners and translations are free. It runs until stop_broadcast or the stream ends.\nTranscriptions running at once per account: 5 on Pay as you go, 50 on Premium; past that a start is refused with quota_exceeded.\nBehaviour: each call starts a new run unless it repeats an idempotencyKey; omitting broadcastId makes a new broadcast with a new link.\nReturns: broadcastId, name, link (anyone who opens it reads, or with spokenTranslations also hears, the stream in their language), transcriptionId, state.\nErrors: invalid_state (with link) when already on air, invalid_options for a non-http(s) address, insufficient_credit (with topUpUrl), quota_exceeded, not_found for an unknown broadcastId, forbidden for a key without the write and recordings:publish scopes.",
538
+ "description": "Start live transcription and translation of a stream playing now, on a public link listeners open in their own language.\nUse: a stream playing now. A recording that has ended: create_transcription. Reading it while it runs: get_transcription.\nArguments: streamUrl must be live when called; the same streamUrl on the same broadcast reuses its stream. broadcastId (from list_broadcasts) reruns an existing broadcast on its old link, and then name is ignored; without it a new broadcast is made, named name or the stream's host. spokenLanguage is a code like en (omit to detect). spokenTranslations true adds audio of each translation.\nList price: $0.60 an hour while it runs, billed per second; Pay as you go bills at least 5 minutes per transcription. Listeners and translations are free.\nTranscriptions running at once per account: 5 on Pay as you go, 50 on Premium; past that a start is refused with quota_exceeded.\nBehaviour: not idempotent (idempotentHint false): every call starts a new billed run, and without broadcastId a new broadcast with a new link. The one exception is a retry: repeating the same arguments with the same idempotencyKey within 24 h returns the first run (replayed true) instead of starting another. A run lasts until stop_broadcast or the stream ends, and billing stops at that moment.\nReturns: broadcastId, name, link (anyone who opens it reads, or with spokenTranslations also hears, the stream in their language), transcriptionId, state.\nErrors: invalid_state (with link) when already on air, invalid_options for a non-http(s) address, insufficient_credit (with topUpUrl), quota_exceeded, not_found for an unknown broadcastId, forbidden for a key without the write and recordings:publish scopes.",
580
539
  "inputSchema": {
581
540
  "additionalProperties": false,
582
541
  "properties": {
@@ -643,7 +602,7 @@
643
602
  {
644
603
  "name": "update_broadcast",
645
604
  "title": "Update a broadcast",
646
- "description": "Change a broadcast's settings: its name, who can listen (public, or private with a password), the language spoken, and whether translations are read aloud, in stock or the speakers' own cloned voices.\nUse: before or between runs; settings apply from the next start_broadcast, the name and access at once. Only the fields given change; give at least one. Starting it: start_broadcast; ending it: stop_broadcast; removing it: delete_broadcast.\nArguments: broadcastId (required) from list_broadcasts or start_broadcast. name 1 to 120 characters. access private needs a password (at least 4 characters) the first time; a new password replaces the old one. spokenLanguage a code like en, or auto. spokenTranslations true or false. clonedVoices true needs consent true, recording that everyone speaking agreed to be cloned.\nBehaviour: free; the link stays the same. Changing access or the password signs out every current listener, who must open the link again. Calling again with the same fields changes nothing else.\nReturns: the broadcast (broadcastId, name, link, access, onAir, listenersNow) with spokenLanguage, spokenTranslations and clonedVoices.\nErrors: invalid_options for nothing to change, a bad name, a short password, private without a password, an unknown language, or clonedVoices true without consent; invalid_state for a broadcast in the trash; not_found for an unknown broadcastId; forbidden for a read-only key.",
605
+ "description": "Change a broadcast's settings: its name, who can listen (public, or private with a password), the language spoken, and whether translations are read aloud, in stock or the speakers' own cloned voices; or bring it back from the trash.\nUse: before or between runs; settings apply from the next start_broadcast, the name and access at once. Only the fields given change; give at least one. Starting it: start_broadcast; ending it: stop_broadcast; removing it: delete_broadcast.\nArguments: broadcastId (required); the rest are independent and optional, except that password counts only with access private (required the first time it goes private) and clonedVoices only with spokenTranslations on, needing consent true. restore true brings a deleted broadcast back first, then applies the rest. Examples: {broadcastId, access: \"private\", password: \"choir2026\"}; {broadcastId, spokenTranslations: true, clonedVoices: true, consent: true}; {broadcastId, restore: true}.\nBehaviour: free; the link and the broadcastId stay the same. Changing access or the password signs out every current listener at once, even on air: their page asks for the link again (and the new password when private), and they miss the lines spoken until they are back; the transcript itself is kept. Language and voice settings never touch a run in progress. Two calls at once: the later one wins, field by field. Works with a key or token of this account with the write scope; a read-only key is refused before anything changes. restore works for 30 days after delete_broadcast; after that the broadcast is not_found. Calling again with the same fields changes nothing else, and rate_limited (with retryAfterS) past 600 calls a minute per API key.\nReturns: the broadcast (broadcastId, name, link, access, onAir, listenersNow) with spokenLanguage, spokenTranslations and clonedVoices, and restored with restore.\nErrors: invalid_options for nothing to change, a bad name, a short password, private without a password, an unknown language, restore false, or clonedVoices true without consent; invalid_state for a broadcast in the trash without restore true; not_found for an unknown broadcastId; forbidden for a read-only key.",
647
606
  "inputSchema": {
648
607
  "additionalProperties": false,
649
608
  "properties": {
@@ -689,6 +648,13 @@
689
648
  "minLength": 4,
690
649
  "type": "string"
691
650
  },
651
+ "restore": {
652
+ "description": "true: bring it back from the trash (after delete_broadcast, within the restore window) with its link, settings and recorded transcriptions; other fields given apply after. false is not accepted.",
653
+ "examples": [
654
+ true
655
+ ],
656
+ "type": "boolean"
657
+ },
692
658
  "spokenLanguage": {
693
659
  "description": "Code of the language spoken (e.g. en, es), or auto to detect it. Applies from the next start_broadcast.",
694
660
  "examples": [
@@ -747,7 +713,7 @@
747
713
  {
748
714
  "name": "delete_broadcast",
749
715
  "title": "Delete a broadcast",
750
- "description": "Move a broadcast to the trash: its public link and stream addresses stop working for listeners.\nUse: a broadcast that will not run again. For 30 days it can be restored from the trash in the VocalStack Dashboard (list_broadcasts with lifecycle archived lists it); after that the broadcast is erased for good. Ending a live run: stop_broadcast.\nArguments: broadcastId from list_broadcasts or start_broadcast; not a transcriptionId.\nBehaviour: free. A broadcast on air is refused: stop it first (stop_broadcast). The transcriptions it recorded are not deleted: while it is in the trash they leave list_transcriptions with it, get_transcription still reads them by id, and delete_transcription deletes one. Deleting again changes nothing.\nReturns: broadcastId, name, deleted true, deletedAt, restorableUntil.\nErrors: invalid_state while on air; not_found for an unknown broadcastId; forbidden for a read-only key.",
716
+ "description": "Move a broadcast to the trash: its public link and stream addresses stop working for listeners.\nUse: a broadcast that will not run again. For 30 days update_broadcast with restore true brings it back with the same link (list_broadcasts with lifecycle archived lists the trash); after that the broadcast is erased for good. Ending a live run: stop_broadcast.\nArguments: broadcastId, a 36-character UUID (8-4-4-4-12 hex digits) from list_broadcasts or start_broadcast; not a transcriptionId (delete_transcription removes its recordings) and not the short code in its link, both of which are not_found. One already in the trash is left as it is, so a retry is safe.\nBehaviour: free, and a soft delete: nothing is erased for 30 days. At once every connected listener is signed out and its link and stream addresses answer not found. A broadcast on air is refused: stop it first (stop_broadcast). The transcriptions it recorded are not deleted: while it is in the trash they leave list_transcriptions and search_transcriptions with it, get_transcription still reads them by id, and delete_transcription deletes one. Restoring brings back the same link and its transcriptions.\nReturns: broadcastId, name, deleted true, deletedAt, restorableUntil.\nErrors: invalid_state while on air; not_found for an unknown broadcastId; forbidden for a read-only key.",
751
717
  "inputSchema": {
752
718
  "additionalProperties": false,
753
719
  "properties": {
@@ -772,6 +738,23 @@
772
738
  "idempotentHint": true,
773
739
  "openWorldHint": false
774
740
  }
741
+ },
742
+ {
743
+ "name": "get_account",
744
+ "title": "Check credit and usage",
745
+ "description": "Check the account's credit and usage: the balance left, the plan, whether a new file or live transcription would start now (and what would refuse it), and this period's limits.\nUse: before a long or live transcription, after an insufficient_credit or quota_exceeded error, or to tell the user what they have left. Prices per minute are in create_transcription and start_broadcast; adding credit happens at topUpUrl, in a browser.\nArguments: none.\nBehaviour: free and read-only; reflects charges as they land, including a live run still being billed. Gives rate_limited (with retryAfterS) past 600 calls a minute per API key.\nReturns: billingEnabled; plan (id, name, kind) or null; balance in US dollars (negative while a plan lets a running transcription continue below zero); canStart and startBlock (no_credit or hourly_limit, null when it starts) for a file; canStartLive and startBlockLive (also concurrency) for a live one; running, concurrentLimit, createdLastHour and createLimit (null means unlimited); period (start, end); monthlyCredit when a plan grants one; usageCap (amount, period, used, resetsAt) when this API key has a spending cap; topUpUrl.\nErrors: rate_limited (with retryAfterS); forbidden for a token without the read scope.",
746
+ "inputSchema": {
747
+ "additionalProperties": false,
748
+ "properties": {},
749
+ "type": "object"
750
+ },
751
+ "annotations": {
752
+ "title": "Check credit and usage",
753
+ "readOnlyHint": true,
754
+ "destructiveHint": false,
755
+ "idempotentHint": true,
756
+ "openWorldHint": false
757
+ }
775
758
  }
776
759
  ]
777
760
  }
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@vocalstack/mcp",
3
- "version": "1.5.0",
3
+ "version": "1.5.2",
4
4
  "description": "VocalStack's MCP server: transcribe, translate, search and summarize audio and video, and run live broadcasts, from any MCP client, on VocalStack's public API.",
5
5
  "license": "MIT",
6
6
  "type": "module",