voxflow 1.15.6 → 1.17.0

package/lib/core/ffmpeg.js

@@ -2,25 +2,60 @@
  * VoxFlow CLI — FFmpeg utility layer
  *
  * Provides ffmpeg/ffprobe operations via child_process.execFile.
- * Zero external dependencies — requires ffmpeg to be installed on PATH.
- * Also provides a one-time dependency hint for users without FFmpeg.
+ * Prefers system FFmpeg on PATH; falls back to `ffmpeg-static` (npm)
+ * when available, so users can `npm install` without system dependency.
  */
 
 const { execFile } = require('child_process');
 const path = require('path');
 const fs = require('fs');
 
+// ── FFmpeg binary resolution ────────────────────────────────────────────────
+
+let _resolvedFfmpegPath = null;
+
+/**
+ * Resolve the ffmpeg binary path. Priority:
+ *   1. System `ffmpeg` on PATH
+ *   2. `ffmpeg-static` npm package (optional dependency)
+ */
+function resolveFfmpegBin() {
+  if (_resolvedFfmpegPath) return _resolvedFfmpegPath;
+
+  // Try system ffmpeg first (faster, no extra dep)
+  try {
+    require('child_process').execFileSync('ffmpeg', ['-version'], { stdio: 'ignore' });
+    _resolvedFfmpegPath = 'ffmpeg';
+    return _resolvedFfmpegPath;
+  } catch { /* not on PATH */ }
+
+  // Fall back to ffmpeg-static
+  try {
+    const staticPath = require('ffmpeg-static');
+    if (staticPath && fs.existsSync(staticPath)) {
+      _resolvedFfmpegPath = staticPath;
+      return _resolvedFfmpegPath;
+    }
+  } catch { /* not installed */ }
+
+  // Nothing found — return 'ffmpeg' and let it fail with a helpful error
+  _resolvedFfmpegPath = 'ffmpeg';
+  return _resolvedFfmpegPath;
+}
+
 /**
  * Run a command via execFile and return { stdout, stderr }.
  * Rejects on non-zero exit code or spawn error.
+ * Automatically resolves 'ffmpeg' to the best available binary.
  * @param {string} cmd
  * @param {string[]} args
  * @param {object} [opts]
  * @returns {Promise<{stdout: string, stderr: string}>}
  */
 function runCommand(cmd, args, opts) {
+  const resolvedCmd = cmd === 'ffmpeg' ? resolveFfmpegBin() : cmd;
   return new Promise((resolve, reject) => {
-    execFile(cmd, args, { timeout: 600_000, ...opts }, (error, stdout, stderr) => {
+    execFile(resolvedCmd, args, { timeout: 600_000, ...opts }, (error, stdout, stderr) => {
       if (error) {
         error.stderr = stderr;
         error.stdout = stdout;
@@ -34,13 +69,16 @@ function runCommand(cmd, args, opts) {
 
 /**
  * Check if ffmpeg and ffprobe are available.
- * @returns {Promise<{available: boolean, version?: string, ffprobeAvailable?: boolean}>}
+ * Uses the auto-resolved binary path (system or ffmpeg-static).
+ * @returns {Promise<{available: boolean, version?: string, ffprobeAvailable?: boolean, source?: string}>}
  */
 async function checkFfmpeg() {
+  const bin = resolveFfmpegBin();
   try {
     const { stdout } = await runCommand('ffmpeg', ['-version']);
     const versionMatch = stdout.match(/ffmpeg version (\S+)/);
     const version = versionMatch ? versionMatch[1] : 'unknown';
+    const source = bin === 'ffmpeg' ? 'system' : 'ffmpeg-static';
 
     // Also check ffprobe
     let ffprobeOk = false;
@@ -55,6 +93,7 @@ async function checkFfmpeg() {
       available: true,
       version,
       ffprobeAvailable: ffprobeOk,
+      source,
     };
   } catch {
     return { available: false };
@@ -363,19 +402,43 @@ async function replaceAudio(videoInput, audioInput, output) {
  * @param {string[]} inputs - Array of input video file paths
  * @param {string} output - Output concatenated video path
  */
-async function concatVideos(inputs, output) {
+async function concatVideos(inputs, output, opts = {}) {
+  const { remux = false } = opts;
   const tmpDir = path.dirname(output);
   const listFile = path.join(tmpDir, `concat-list-${Date.now()}.txt`);
   const content = inputs.map(f => `file '${path.resolve(f)}'`).join('\n');
   fs.writeFileSync(listFile, content, 'utf8');
 
   try {
-    await runCommand('ffmpeg', [
-      '-f', 'concat', '-safe', '0',
-      '-i', listFile,
-      '-c', 'copy',
-      '-y', path.resolve(output),
-    ]);
+    if (remux) {
+      // Two-pass: concat then remux to strip AAC priming edit list
+      // (fixes macOS QuickTime showing black on first frame)
+      const rawOutput = path.resolve(output) + '.tmp.mp4';
+      await runCommand('ffmpeg', [
+        '-f', 'concat', '-safe', '0',
+        '-i', listFile,
+        '-c', 'copy',
+        '-movflags', '+faststart',
+        '-y', rawOutput,
+      ]);
+      await runCommand('ffmpeg', [
+        '-ss', '0',
+        '-i', rawOutput,
+        '-c', 'copy',
+        '-movflags', '+faststart',
+        '-y', path.resolve(output),
+      ]);
+      try { fs.unlinkSync(rawOutput); } catch {}
+    } else {
+      // Single-pass concat (default, used by picstory/present/summarize)
+      await runCommand('ffmpeg', [
+        '-f', 'concat', '-safe', '0',
+        '-i', listFile,
+        '-c', 'copy',
+        '-movflags', '+faststart',
+        '-y', path.resolve(output),
+      ]);
+    }
   } finally {
     try { fs.unlinkSync(listFile); } catch {}
   }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "voxflow",
-  "version": "1.15.6",
+  "version": "1.17.0",
   "description": "AI audio content creation CLI — stories, podcasts, narration, dubbing, transcription, translation, and video translation with TTS",
   "bin": {
     "voxflow": "./dist/index.js"
@@ -51,7 +51,7 @@
     "lint:fix": "eslint lib/ bin/ tests/ --fix",
     "test": "node --test tests/*.test.js",
     "test:watch": "node --test --watch tests/*.test.js",
-    "version": "node scripts/sync-plugin-versions.js && git add .claude-plugin/plugin.json .cursor-plugin/plugin.json .codex-plugin/plugin.json"
+    "version": "node scripts/sync-plugin-versions.js && git add .claude-plugin/plugin.json .cursor-plugin/plugin.json .codex-plugin/plugin.json skills/.claude-plugin/plugin.json"
   },
   "author": "gonghaoran",
   "homepage": "https://voxflow.studio",
@@ -68,5 +68,8 @@
     "@remotion/bundler": "4.0.448",
     "@vercel/ncc": "^0.38.4",
     "eslint": "^9.39.4"
+  },
+  "optionalDependencies": {
+    "ffmpeg-static": "^5.3.0"
   }
 }

package/skills/.claude-plugin/marketplace.json

@@ -1,7 +1,7 @@
 {
   "$schema": "https://anthropic.com/claude-code/marketplace.schema.json",
   "name": "voxflow",
-  "description": "VoxFlow — AI voice CLI as a skill for any AI agent. Bundles 5 skills (hub, podcast, transcribe, video, slice) that drive a hosted TTS / ASR / podcast / render backend.",
+  "description": "VoxFlow — AI voice CLI as a skill for any AI agent. Bundles 6 skills (hub, podcast, transcribe, video, slice, card) that drive a hosted TTS / ASR / podcast / render backend.",
   "owner": {
     "name": "VoxFlow",
     "url": "https://voxflow.studio"
@@ -9,7 +9,7 @@
   "plugins": [
     {
       "name": "voxflow",
-      "description": "AI voice CLI bundled as 5 skills. TTS in 200+ voices across 40+ languages, multi-speaker AI podcasts, audio/video transcription with word-level timestamps, SRT-driven dubbing, end-to-end video translation, and long-article → vertical card video reels via Remotion. Backed by a hosted TTS/ASR/LLM/render service with per-user quota (free tier 10K/mo).",
+      "description": "AI voice CLI bundled as 6 skills. TTS in 200+ voices across 40+ languages, multi-speaker AI podcasts, audio/video transcription with word-level timestamps, SRT-driven dubbing, end-to-end video translation, long-article → vertical card video reels via Remotion, and turning text into polished shareable card images or narrated card videos. Backed by a hosted TTS/ASR/LLM/render service with per-user quota (free tier 10K/mo).",
       "category": "productivity",
       "source": "./",
       "homepage": "https://voxflow.studio",

package/skills/.claude-plugin/plugin.json

@@ -1,7 +1,7 @@
 {
   "name": "voxflow",
-  "version": "1.15.0",
-  "description": "AI voice CLI bundled as 5 skills (hub, podcast, transcribe, video, slice). Synthesize speech in 200+ voices across 40+ languages, generate multi-speaker AI podcasts, transcribe audio/video with word-level timestamps, dub videos from SRT subtitles, run end-to-end video translation, and turn long articles into vertical card video reels via Remotion. Backed by a hosted TTS/ASR/LLM/render service with per-user quota (free tier 10K/mo).",
+  "version": "1.17.0",
+  "description": "AI voice CLI bundled as 6 skills (hub, podcast, transcribe, video, slice, card). Synthesize speech in 200+ voices across 40+ languages, generate multi-speaker AI podcasts, transcribe audio/video with word-level timestamps, dub videos from SRT subtitles, run end-to-end video translation, turn long articles into vertical card video reels via Remotion, and turn text into polished shareable card images or narrated card videos. Backed by a hosted TTS/ASR/LLM/render service with per-user quota (free tier 10K/mo).",
   "author": {
     "name": "VoxFlow",
     "url": "https://voxflow.studio"
@@ -19,6 +19,7 @@
     "dubbing",
     "video",
     "remotion",
+    "card",
     "voxflow",
     "skill"
   ]

package/skills/README.md

@@ -2,12 +2,12 @@
 
 [![License](https://img.shields.io/badge/license-Apache%202.0-blue.svg)](LICENSE)
 [![npm version](https://img.shields.io/npm/v/voxflow.svg)](https://www.npmjs.com/package/voxflow)
-[![5 skills](https://img.shields.io/badge/skills-5-purple.svg)](#what-you-get)
+[![6 skills](https://img.shields.io/badge/skills-6-purple.svg)](#what-you-get)
 [![Works with](https://img.shields.io/badge/works%20with-Claude%20Code%20%C2%B7%20Cursor%20%C2%B7%20Codex%20%C2%B7%20Gemini-orange.svg)](#install)
 
 ![VoxFlow CLI demo](https://github.com/VoxFlowStudio/skills/raw/main/demo.gif)
 
-Voice in your AI workflow. Five skills that let any AI coding agent (Claude Code · Cursor · Codex · Gemini CLI · Cline) speak in 200+ voices, generate podcasts, dub videos, and transcribe audio — through one CLI.
+Voice in your AI workflow. Six skills that let any AI coding agent (Claude Code · Cursor · Codex · Gemini CLI · Cline) speak in 200+ voices, generate podcasts, dub videos, transcribe audio, and turn text into card images or narrated card videos — through one CLI.
 
 **Why VoxFlow over a raw TTS API?** One CLI handles auth, voice search, multi-speaker dialogue, video pipelines, and quota. The skills layer makes it native to whichever agent you're already using — no new context-switch.
 
@@ -30,8 +30,8 @@ npx -y skills add VoxFlowStudio/skills --all --yes --global
 
 The `skills` npm package detects every AI agent on your machine (Claude Code,
 Cursor, Codex CLI, Gemini CLI, Cline, Amp, Antigravity, CodeBuddy, OpenClaw…)
-and writes the 5 VoxFlow skills (`hub`, `podcast`, `transcribe`, `video`,
-`slice`) to each agent's standard skill location in a single shot.
+and writes the 6 VoxFlow skills (`hub`, `podcast`, `transcribe`, `video`,
+`slice`, `card`) to each agent's standard skill location in a single shot.
 
 ## Prerequisites
 
@@ -42,7 +42,7 @@ voxflow login          # one-time browser auth
 
 ## What You Get
 
-Five focused skills, each loaded on demand:
+Six focused skills, each loaded on demand:
 
 | Skill | Invoked as | What it covers |
 |-------|-----------|----------------|
@@ -50,7 +50,8 @@ Five focused skills, each loaded on demand:
 | **podcast** | `voxflow:podcast` | Multi-speaker AI podcast from topic / URL / script |
 | **transcribe** | `voxflow:transcribe` | `asr` · `asr-jobs` · `translate` · `dub` · `video-translate` · `summarize` · `publish` |
 | **video** | `voxflow:video` | `picstory` · `present` · `slides` · `explain` · `image` |
-| **slice** | `voxflow:slice` | Article → vertical card video (1080×1920); 6 themes: paper / editorial / poster / Notion / brutalist / glass |
+| **slice** | `voxflow:slice` | Article → vertical card video (1080×1920); 13 editorial / poster / magazine themes |
+| **card** | `voxflow:card` | Text → shareable card images (HTML/CSS + Playwright); 1:1 / 3:4 / 9:16, editorial design system. Optional narrated MP4 video via `voxflow card render` (TTS + FFmpeg, in-project output) |
 
 Use the hub skill as the starting point — it routes to the others automatically.
 
@@ -76,7 +77,8 @@ voxflow/
   podcast/SKILL.md      # AI dialogue podcast
   transcribe/SKILL.md   # ASR, translation, dubbing
   video/SKILL.md        # AI short video, slides, images
-  slice/SKILL.md        # Article → vertical card video (6 themes)
+  slice/SKILL.md        # Article → vertical card video (13 themes)
+  card/SKILL.md         # Text → shareable card images (1:1 / 3:4 / 9:16)
 ```
 
 ## Registry