@pinecall/skills 0.1.13 → 0.1.15

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pinecall/skills",
-  "version": "0.1.13",
+  "version": "0.1.15",
   "description": "Agent Skills for the Pinecall SDK — installable into Claude Code, Antigravity, Cursor, Copilot and any agent that supports the open Skills format.",
   "type": "module",
   "license": "MIT",

package/skills/pinecall-reference/references/reference/managed-vs-byok.md

@@ -34,12 +34,12 @@ Every STT, TTS and LLM model on Pinecall is one of two kinds:
 
 | Service | BYOK-only providers |
 |---|---|
-| **STT** | `assemblyai`, `soniox` |
+| **STT** | `assemblyai`, `soniox`, `xai` (Grok) |
 | **TTS** | `rime`, `soniox`, `xai` (Grok voices) |
 | **LLM** | `xai` (grok), `groq`, `cerebras`, `deepseek`, `openrouter` |
 
 > One key covers multiple services: a **Soniox** key = STT + TTS; an **xAI** key =
-> Grok LLM + Grok TTS voices.
+> Grok **LLM + TTS + STT**.
 
 ## Check it from the API (authoritative, live)
 

package/skills/pinecall-reference/references/reference/stt-providers.md

@@ -30,6 +30,7 @@ Pinecall supports multiple STT providers. Use the `provider/model` format or a f
 { stt: "elevenlabs/scribe" }         // ElevenLabs Scribe v2 (realtime)
 { stt: "assemblyai/universal" }      // AssemblyAI Universal-3
 { stt: "soniox/realtime" }           // Soniox real-time (BYOK)
+{ stt: "xai/grok-stt" }              // xAI Grok STT (BYOK)
 ```
 
 ## Managed vs bring-your-own-key (BYOK)
@@ -48,6 +49,7 @@ for the full list and the live `GET /api/rates/models` query.
 | `elevenlabs` (scribe) | ✅ Yes | Same key as ElevenLabs TTS — Pinecall hosts it |
 | `assemblyai` (universal) | ❌ BYOK only | Add an AssemblyAI key |
 | `soniox` (realtime) | ❌ BYOK only | One Soniox key = STT **and** TTS |
+| `xai` (grok-stt) | ❌ BYOK only | Same xAI key as Grok LLM/TTS |
 
 > **BYOK enforcement:** if you configure a BYOK-only STT provider and your org has
 > not saved a key for it, **agent registration is rejected** with
@@ -96,7 +98,7 @@ Or with tuning:
 ```typescript
 stt: {
   provider: "deepgram",
-  model: "nova-3",
+  model: "nova-3",            // "nova-3" | "nova-2"
   language: "en",
   interim_results: true,
   smart_format: true,
@@ -104,7 +106,9 @@ stt: {
   profanity_filter: false,
   endpointing_ms: 300,
   utterance_end_ms: 1000,
-  keywords: ["pinecall"],
+  keywords: ["pinecall"],      // nova-2 keyword boosting
+  keyterms: ["pinecall"],      // nova-3 keyterm prompting
+  min_confidence: 0.0,         // drop transcripts below this confidence (0 = off)
 }
 ```
 
@@ -192,6 +196,16 @@ stt: "soniox/realtime"
 stt: { provider: "soniox", model: "stt-rt-v5", language: "en" }
 ```
 
+## xAI Grok (BYOK)
+
+Grok speech-to-text — same **xAI key** as Grok LLM and TTS. Requires your own key.
+
+```typescript
+stt: "xai/grok-stt"
+// or
+stt: { provider: "xai", model: "grok-stt", language: "en" }
+```
+
 ## Which to choose
 
 | Provider | Best for | Trade-off |
@@ -204,6 +218,7 @@ stt: { provider: "soniox", model: "stt-rt-v5", language: "en" }
 | `elevenlabs/scribe` | Single-vendor with ElevenLabs TTS | Managed (shared key) |
 | `assemblyai/universal` | Accuracy + diarization | BYOK only |
 | `soniox/realtime` | Multilingual (60+), single-vendor with Soniox TTS | BYOK only |
+| `xai/grok-stt` | Single-vendor with Grok LLM + TTS | BYOK only |
 
 For most agents, start with `deepgram/flux`. Use `deepgram/nova-3` for languages Flux doesn't cover (Arabic, Hindi, Thai, Chinese, Japanese, Korean, etc.).
 

package/skills/pinecall-reference/references/reference/tts-providers.md

@@ -179,8 +179,11 @@ Shortcut: `"cartesia/yumiko"`
 voice: {
   provider: "polly",
   voice_id: "Joanna",
-  engine: "neural",
+  engine: "neural",        // "neural" | "standard"
   language: "en-US",
+  rate: "medium",          // "slow" | "medium" | "fast" | "+10%" / "-10%"
+  volume: "medium",        // "soft" | "medium" | "loud" | "+6dB" / "-6dB"
+  pitch: "+5%",            // standard engine only
 }
 ```
 
@@ -189,6 +192,7 @@ Shortcut: `"polly/joanna"`
 **Tuning notes:**
 
 - `engine: "neural"` is required for natural-sounding output. The older `standard` engine is robotic.
+- `rate` / `volume` accept named levels or relative values; `pitch` only applies to the `standard` engine.
 - Polly is the cheapest option but the least natural — fine for IVR-style flows, not for engaging conversation.
 
 ## Rime (BYOK)