@mrgoonie/multix 0.12.0 → 0.14.0

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@mrgoonie/multix",
-  "version": "0.12.0",
-  "description": "AI multimodal CLI — OpenAI, Gemini, MiniMax, OpenRouter, Leonardo, BytePlus, ElevenLabs, fal.ai, Atlas Cloud, Replicate, ffmpeg, ImageMagick, doc-to-md",
+  "version": "0.14.0",
+  "description": "AI multimodal CLI — OpenAI, Gemini, MiniMax, OpenRouter, Leonardo, BytePlus, ElevenLabs, fal.ai, Atlas Cloud, Replicate, HeyGen, ffmpeg, ImageMagick, doc-to-md",
   "type": "module",
   "bin": {
     "multix": "./dist/cli.js"
@@ -51,6 +51,9 @@
     "fal-ai",
     "atlascloud",
     "replicate",
+    "heygen",
+    "ai-avatar",
+    "video-translation",
     "ffmpeg",
     "imagemagick",
     "media",

package/skill/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: multix
-description: AI multimodal CLI — generate and edit images (OpenAI, Codex driver, Gemini Nano Banana, Imagen, MiniMax, OpenRouter, Leonardo, BytePlus Seedream, Cloudflare Workers AI FLUX.1 Schnell, fal.ai FLUX, Replicate FLUX), generate video (Gemini Veo, MiniMax Hailuo, OpenRouter via Veo/Seedance/Kling, Leonardo via Motion/Veo/Kling/Hailuo/Seedance/LTX, BytePlus Seedance, Cloudflare AI Gateway via Replicate, fal.ai queue models, Replicate direct API models incl. P-Video), TTS/STT/music (OpenAI, Gemini Flash TTS incl. Gemini 3.8 with delivery-style control, MiniMax, Cloudflare Workers AI MeloTTS, ElevenLabs), 3D models (BytePlus Hyper3D/Hitem3d), analyze/transcribe media, convert documents to Markdown, optimize media via ffmpeg/ImageMagick, and self-update the CLI. Use whenever the user wants to create, edit, or transform images/video/audio/3D/documents from a CLI, mentions any of: OpenAI, Codex image driver, image-to-image, i2i, image edit, watercolor/cyberpunk style transfer, reference image, OpenRouter, Nano Banana, Flux, Seedream, Veo, Hailuo, Seedance, Kling, Cloudflare Workers AI, AI Gateway, fal.ai, Atlas Cloud, Replicate, replicate.com models, Replicate predictions, video generation, text-to-video, image-to-video, i2v, ElevenLabs voice cloning, updating/upgrading multix, or asks for batch media optimization.
-version: 0.12.0 # x-release-please-version
+description: AI multimodal CLI — generate and edit images (OpenAI, Codex driver, Gemini Nano Banana, Imagen, MiniMax, OpenRouter, Leonardo, BytePlus Seedream, Cloudflare Workers AI FLUX.1 Schnell, fal.ai FLUX, Replicate FLUX), generate video (Gemini Veo, MiniMax Hailuo, OpenRouter via Veo/Seedance/Kling, Leonardo via Motion/Veo/Kling/Hailuo/Seedance/LTX, BytePlus Seedance, Cloudflare AI Gateway via Replicate, fal.ai queue models, Replicate direct API models incl. P-Video, HeyGen Video Agent and scripted avatar videos), video translation and lipsync (HeyGen), TTS/STT/music (OpenAI, Gemini Flash TTS incl. Gemini 3.8 with delivery-style control, MiniMax, Cloudflare Workers AI MeloTTS, ElevenLabs), 3D models (BytePlus Hyper3D/Hitem3d), analyze/transcribe media, convert documents to Markdown, optimize media via ffmpeg/ImageMagick, and self-update the CLI. Use whenever the user wants to create, edit, or transform images/video/audio/3D/documents from a CLI, mentions any of: OpenAI, Codex image driver, image-to-image, i2i, image edit, watercolor/cyberpunk style transfer, reference image, OpenRouter, Nano Banana, Flux, Seedream, Veo, Hailuo, Seedance, Kling, Cloudflare Workers AI, AI Gateway, fal.ai, Atlas Cloud, Replicate, replicate.com models, Replicate predictions, HeyGen, avatar video, talking head, Video Agent, video translation, dubbing, lipsync, video generation, text-to-video, image-to-video, i2v, ElevenLabs voice cloning, updating/upgrading multix, or asks for batch media optimization.
+version: 0.14.0 # x-release-please-version
 ---
 
 # multix — AI Multimodal CLI
@@ -24,6 +24,7 @@ Activate this skill when the user asks to:
 - Run any fal.ai model (generic queue-API runner), or fal.ai's default image/video models
 - Run any Atlas Cloud model (image, image edit, video, TTS/STT, 3D) with schema-validated inputs and resumable async predictions
 - Run any Replicate model (official or versioned) via the direct API, generate Replicate images/video, or resume/cancel a Replicate prediction by id
+- Make HeyGen avatar videos from a prompt (Video Agent) or an exact script, translate/lipsync videos, HeyGen TTS
 - Update multix itself to the latest (or a given dist-tag) npm version
 
 ## Self-update
@@ -47,6 +48,7 @@ ELEVENLABS_API_KEY    # ElevenLabs (TTS, cloning, STT, music, dubbing)
 FAL_KEY                # fal.ai (generic queue-API model runner, image/video)
 ATLASCLOUD_API_KEY    # Atlas Cloud (any model: image, edit, video, audio, 3D)
 REPLICATE_API_TOKEN   # Replicate direct API (REPLICATE_API_KEY accepted as fallback)
+HEYGEN_API_KEY        # HeyGen v3 (avatar video, Video Agent, translation); API billed separately from web plans
 CLOUDFLARE_ACCOUNT_ID # Cloudflare Workers AI / AI Gateway account
 CLOUDFLARE_API_TOKEN  # Cloudflare Workers AI / AI Gateway token
 # Video additionally requires CLOUDFLARE_AI_GATEWAY_ID and REPLICATE_API_TOKEN.
@@ -178,6 +180,32 @@ multix replicate cancel <prediction-id>
 
 Shared flags: `--async`, `--wait`, `--prefer-wait [seconds]` (sends `Prefer: wait=N`, still polls if unfinished), `--wait-timeout <ms>`, `--no-download` (print URLs), `--output <path|dir/>`. `failed`/`canceled`/`aborted` predictions exit non-zero. Local files up to 256 KB are inlined as data URIs; larger files (< 100 MB) go through the Replicate Files API.
 
+## HeyGen
+
+Auth is `X-Api-Key: $HEYGEN_API_KEY` (v3 API). API credits are billed separately from HeyGen web plans; check with `multix heygen me`.
+
+Pick the command by who writes the script:
+- `video-agent "<prompt>"`: **Video Agent**, prompt-to-video. The agent writes the script and picks avatar/voice/scenes. Best for drafts and explainers from a brief.
+- `video "<script>" --avatar <look-id>`: **scripted avatar video**. The avatar speaks your exact words (or lip-syncs `--audio`). Get look IDs from `heygen avatars`, not group IDs.
+- `model-video "<prompt>"`: the HeyGen Video model, a 5-15 s cinematic clip with no avatar.
+
+```bash
+multix heygen video-agent "45-second explainer for a note app" [--avatar <look-id>] [--voice <voice-id>] [--style <style-id>] [--orientation portrait] [--file <ref>]
+multix heygen avatars [--type photo_avatar] [--cursor <token>]
+multix heygen voices [--language English]
+multix heygen video "Hello!" --avatar <look-id> [--voice <id>] [--aspect-ratio 9:16] [--engine avatar_iv|avatar_v|avatar_iii] [--burn-captions]
+multix heygen translate <ref> --languages "Spanish, Chinese (Cantonese, Traditional)" [--output <dir|dub.mp4>]   # several languages: dub.mp4 -> dub-Spanish.mp4, ...
+multix heygen lipsync --video <ref> --audio <ref>
+multix heygen tts "Hello" --voice <voice-id>
+multix heygen template <template-id> --input @vars.json
+multix heygen run <video-agent|video|video-translation|lipsync|model-video|speech> --input @payload.json
+multix heygen status <id> --type agent|video|translation|lipsync|model-video --wait --download
+```
+
+- `<ref>` = `https://` URL, `asset:<asset_id>`, or a local file (uploaded, max 32 MB).
+- Commands wait and download by default. `--async` prints IDs and a resume command; `--type agent` resumes with the Video Agent `session_id`.
+- Failed jobs exit 1 with HeyGen's failure code. 402 means add API credits; 403 means a scope or entitlement is missing.
+
 ## Image-to-image (edit existing image with a prompt)
 
 Every supported provider exposes `image-to-image` and the alias `i2i`. Refs accept local file path OR URL.
@@ -326,6 +354,7 @@ multix byteplus 3d --input-image ./front.png ./side.png --prompt "Generate from
 
 # Docs → Markdown (Gemini)
 multix doc convert --input report.pdf [--auto-name] [--output ./out.md]
+cat report.pdf | multix doc convert --input - --output ./out.md   # "-" reads stdin; --stdin-ext <ext> (e.g. pdf) if undetected
 
 # Media optimization (ffmpeg + ImageMagick)
 multix media optimize --input <file> --output <file> [--target-size <MB>] [--quality 85] [--max-width 1920]