@pinecall/skills 0.1.12 → 0.1.14

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/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@pinecall/skills",
3
- "version": "0.1.12",
3
+ "version": "0.1.14",
4
4
  "description": "Agent Skills for the Pinecall SDK — installable into Claude Code, Antigravity, Cursor, Copilot and any agent that supports the open Skills format.",
5
5
  "type": "module",
6
6
  "license": "MIT",
@@ -34,11 +34,12 @@ Every STT, TTS and LLM model on Pinecall is one of two kinds:
34
34
 
35
35
  | Service | BYOK-only providers |
36
36
  |---|---|
37
- | **STT** | `assemblyai`, `soniox` |
38
- | **TTS** | `rime`, `soniox` |
37
+ | **STT** | `assemblyai`, `soniox`, `xai` (Grok) |
38
+ | **TTS** | `rime`, `soniox`, `xai` (Grok voices) |
39
39
  | **LLM** | `xai` (grok), `groq`, `cerebras`, `deepseek`, `openrouter` |
40
40
 
41
- > `soniox` is one key for **both** STT and TTS (a Soniox key enables both).
41
+ > One key covers multiple services: a **Soniox** key = STT + TTS; an **xAI** key =
42
+ > Grok **LLM + TTS + STT**.
42
43
 
43
44
  ## Check it from the API (authoritative, live)
44
45
 
@@ -30,6 +30,7 @@ Pinecall supports multiple STT providers. Use the `provider/model` format or a f
30
30
  { stt: "elevenlabs/scribe" } // ElevenLabs Scribe v2 (realtime)
31
31
  { stt: "assemblyai/universal" } // AssemblyAI Universal-3
32
32
  { stt: "soniox/realtime" } // Soniox real-time (BYOK)
33
+ { stt: "xai/grok-stt" } // xAI Grok STT (BYOK)
33
34
  ```
34
35
 
35
36
  ## Managed vs bring-your-own-key (BYOK)
@@ -48,6 +49,7 @@ for the full list and the live `GET /api/rates/models` query.
48
49
  | `elevenlabs` (scribe) | ✅ Yes | Same key as ElevenLabs TTS — Pinecall hosts it |
49
50
  | `assemblyai` (universal) | ❌ BYOK only | Add an AssemblyAI key |
50
51
  | `soniox` (realtime) | ❌ BYOK only | One Soniox key = STT **and** TTS |
52
+ | `xai` (grok-stt) | ❌ BYOK only | Same xAI key as Grok LLM/TTS |
51
53
 
52
54
  > **BYOK enforcement:** if you configure a BYOK-only STT provider and your org has
53
55
  > not saved a key for it, **agent registration is rejected** with
@@ -192,6 +194,16 @@ stt: "soniox/realtime"
192
194
  stt: { provider: "soniox", model: "stt-rt-v5", language: "en" }
193
195
  ```
194
196
 
197
+ ## xAI Grok (BYOK)
198
+
199
+ Grok speech-to-text — same **xAI key** as Grok LLM and TTS. Requires your own key.
200
+
201
+ ```typescript
202
+ stt: "xai/grok-stt"
203
+ // or
204
+ stt: { provider: "xai", model: "grok-stt", language: "en" }
205
+ ```
206
+
195
207
  ## Which to choose
196
208
 
197
209
  | Provider | Best for | Trade-off |
@@ -204,6 +216,7 @@ stt: { provider: "soniox", model: "stt-rt-v5", language: "en" }
204
216
  | `elevenlabs/scribe` | Single-vendor with ElevenLabs TTS | Managed (shared key) |
205
217
  | `assemblyai/universal` | Accuracy + diarization | BYOK only |
206
218
  | `soniox/realtime` | Multilingual (60+), single-vendor with Soniox TTS | BYOK only |
219
+ | `xai/grok-stt` | Single-vendor with Grok LLM + TTS | BYOK only |
207
220
 
208
221
  For most agents, start with `deepgram/flux`. Use `deepgram/nova-3` for languages Flux doesn't cover (Arabic, Hindi, Thai, Chinese, Japanese, Korean, etc.).
209
222
 
@@ -33,6 +33,7 @@ for the full list and the live `GET /api/rates/models` query.
33
33
  | `polly` (AWS) | ✅ Yes | |
34
34
  | `rime` | ❌ BYOK only | Add a Rime key under Provider Keys |
35
35
  | `soniox` | ❌ BYOK only | One Soniox key = TTS **and** STT |
36
+ | `xai` (Grok) | ❌ BYOK only | Same xAI key as Grok LLM |
36
37
 
37
38
  > **BYOK enforcement:** configuring `rime` without a saved Rime key rejects agent
38
39
  > registration with `PROVIDER_KEY_REQUIRED`. With your own key, that usage is billed
@@ -221,6 +222,21 @@ voice: {
221
222
 
222
223
  Shortcut: `"soniox/Adrian"`
223
224
 
225
+ ## xAI Grok (BYOK)
226
+
227
+ Expressive Grok voices: **ara, eve, leo, rex, sal**. Uses the **same xAI key** as
228
+ Grok LLM (`XAI_API_KEY`). Requires your own key.
229
+
230
+ ```typescript
231
+ voice: {
232
+ provider: "xai",
233
+ voice_id: "eve", // ara | eve | leo | rex | sal
234
+ speed: 1.0,
235
+ }
236
+ ```
237
+
238
+ Shortcut: `"xai/eve"`
239
+
224
240
  ## Which to choose
225
241
 
226
242
  | Provider | Best for | Trade-off |
@@ -230,6 +246,7 @@ Shortcut: `"soniox/Adrian"`
230
246
  | **Polly** | Cheap IVR, simple flows | Less natural |
231
247
  | **Rime** | Ultra-natural expressive English | BYOK only; English-focused |
232
248
  | **Soniox** | Multilingual (60+), single-vendor with Soniox STT | BYOK only |
249
+ | **xAI Grok** | Expressive Grok voices (ara/eve/leo/rex/sal) | BYOK only |
233
250
 
234
251
  For most agents, start with ElevenLabs (`eleven_flash_v2_5`) or Cartesia (`sonic-3.5`). Use Polly only for high-volume, low-engagement flows.
235
252