[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6JMAENHYTDAAMHTXGTJ1KM5":3,"skill-same:01M6JMAENHYTDAAMHTXGTJ1KM5":96},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":26,"publishedAt":47,"linkOnly":48,"markdownContent":49,"markdownBody":50,"frontMatter":51,"refName":55,"commitSha":56,"skillDirectory":57,"forksCount":58,"repositoryHtmlUrl":59,"repositoryHomepageUrl":60,"ownerHtmlUrl":61,"ownerBlogUrl":62,"ownerDescription":6,"updatedAt":63,"install":64,"resources":77},"01M6JMAENHYTDAAMHTXGTJ1KM5","audio_transcriber",null,"Transform audio recordings into professional Markdown documentation with intelligent summaries using LLM integration","LongLeo287\u002FOmniClaw","LongLeo287","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F141803367?v=4","01M6JK75C62VCNZKR5BRZ386YV","LongLeo","01M6JK75GQ78TSP1ZPXNKK3MCH",9,"MIT","https:\u002F\u002Fgithub.com\u002FLongLeo287\u002FOmniClaw\u002Fblob\u002Fmain\u002Fecosystem\u002Fskills\u002Faudio_transcriber\u002FSKILL.md",[18,23],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000016","图像与多媒体","SCENARIO","使用场景",{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",[27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46],"Audio","audio transcriber","transcription","summaries","minutes","meeting","with","llm integration","transform","professional","meeting-minutes","safe","community","recordings","whisper","text","markdown","integration","speech","documentation","2026-11-10 10:12:30",false,"---\nname: audio_transcriber\n\ndescription: \"Transform audio recordings into professional Markdown documentation with intelligent summaries using LLM integration\"\ncategory: content\nrisk: safe\nsource: community\ntags: \"[audio, transcription, whisper, meeting-minutes, speech-to-text]\"\ndate_added: \"2026-02-27\"\n---\n\n## Purpose\n\nThis skill automates audio-to-text transcription with professional Markdown output, extracting rich technical metadata (speakers, timestamps, language, file size, duration) and generating structured meeting minutes and executive summaries. It uses Faster-Whisper or Whisper with zero configuration, working universally across projects without hardcoded paths or API keys.\n\nInspired by tools like Plaud, this skill transforms raw audio recordings into actionable documentation, making it ideal for meetings, interviews, lectures, and content analysis.\n\n## When to Use\nInvoke this skill when:\n\n- User needs to transcribe audio\u002Fvideo files to text\n- User wants meeting minutes automatically generated from recordings\n- User requires speaker identification (diarization) in conversations\n- User needs subtitles\u002Fcaptions (SRT, VTT formats)\n- User wants executive summaries of long audio content\n- User asks variations of \"transcribe this audio\", \"convert audio to text\", \"generate meeting notes from recording\"\n- User has audio files in common formats (MP3, WAV, M4A, OGG, FLAC, WEBM)\n\n## Workflow\n\n### Step 0: Discovery (Auto-detect Transcription Tools)\n\n**Objective:** Identify available transcription engines without user configuration.\n\n**Actions:**\n\nRun detection commands to find installed tools:\n\n```bash\n# Check for Faster-Whisper (preferred - 4-5x faster)\nif python3 -c \"import faster_whisper\" 2>\u002Fdev\u002Fnull; then\n    TRANSCRIBER=\"faster-whisper\"\n    echo \"✅ Faster-Whisper detected (optimized)\"\n# Fallback to original Whisper\nelif python3 -c \"import whisper\" 2>\u002Fdev\u002Fnull; then\n    TRANSCRIBER=\"whisper\"\n    echo \"✅ OpenAI Whisper detected\"\nelse\n    TRANSCRIBER=\"none\"\n    echo \"⚠️  No transcription tool found\"\nfi\n\n# Check for ffmpeg (audio format conversion)\nif command -v ffmpeg &>\u002Fdev\u002Fnull; then\n    echo \"✅ ffmpeg available (format conversion enabled)\"\nelse\n    echo \"ℹ️  ffmpeg not found (limited format support)\"\nfi\n```\n\n**If no transcriber found:**\n\nOffer automatic installation using the provided script:\n\n```bash\necho \"⚠️  No transcription tool found\"\necho \"\"\necho \"? Auto-install dependencies? (Recommended)\"\nread -p \"Run installation script? [Y\u002Fn]: \" AUTO_INSTALL\n\nif [[ ! \"$AUTO_INSTALL\" =~ ^[Nn] ]]; then\n    # Get skill directory (works for both repo and symlinked installations)\n    SKILL_DIR=\"$(cd \"$(dirname \"${BASH_SOURCE[0]}\")\" && pwd)\"\n    \n    # Run installation script\n    if [[ -f \"$SKILL_DIR\u002Fscripts\u002Finstall-requirements.sh\" ]]; then\n        bash \"$SKILL_DIR\u002Fscripts\u002Finstall-requirements.sh\"\n    else\n        echo \"❌ Installation script not found\"\n        echo \"\"\n        echo \"? Manual installation:\"\n        echo \"  pip install faster-whisper  # Recommended\"\n        echo \"  pip install openai-whisper  # Alternative\"\n        echo \"  brew install ffmpeg         # Optional (macOS)\"\n        exit 1\n    fi\n    \n    # Verify installation succeeded\n    if python3 -c \"import faster_whisper\" 2>\u002Fdev\u002Fnull || python3 -c \"import whisper\" 2>\u002Fdev\u002Fnull; then\n        echo \"✅ Installation successful! Proceeding with transcription...\"\n    else\n        echo \"❌ Installation failed. Please install manually.\"\n        exit 1\n    fi\nelse\n    echo \"\"\n    echo \"? Manual installation required:\"\n    echo \"\"\n    echo \"Recommended (fastest):\"\n    echo \"  pip install faster-whisper\"\n    echo \"\"\n    echo \"Alternative (original):\"\n    echo \"  pip install openai-whisper\"\n    echo \"\"\n    echo \"Optional (format conversion):\"\n    echo \"  brew install ffmpeg  # macOS\"\n    echo \"  apt install ffmpeg   # Linux\"\n    echo \"\"\n    exit 1\nfi\n```\n\nThis ensures users can install dependencies with one confirmation, or opt for manual installation if preferred.\n\n**If transcriber found:**\n\nProceed to Step 0b (CLI Detection).\n\n\n### Step 1: Validate Audio File\n\n**Objective:** Verify file exists, check format, and extract metadata.\n\n**Actions:**\n\n1. **Accept file path or URL** from user:\n   - Local file: `meeting.mp3`\n   - URL: `https:\u002F\u002Fexample.com\u002Faudio.mp3` (download to temp directory)\n\n2. **Verify file exists:**\n\n```bash\nif [[ ! -f \"$AUDIO_FILE\" ]]; then\n    echo \"❌ File not found: $AUDIO_FILE\"\n    exit 1\nfi\n```\n\n3. **Extract metadata** using ffprobe or file utilities:\n\n```bash\n# Get file size\nFILE_SIZE=$(du -h \"$AUDIO_FILE\" | cut -f1)\n\n# Get duration and format using ffprobe\nDURATION=$(ffprobe -v error -show_entries format=duration \\\n    -of default=noprint_wrappers=1:nokey=1 \"$AUDIO_FILE\" 2>\u002Fdev\u002Fnull)\nFORMAT=$(ffprobe -v error -select_streams a:0 -show_entries \\\n    stream=codec_name -of default=noprint_wrappers=1:nokey=1 \"$AUDIO_FILE\" 2>\u002Fdev\u002Fnull)\n\n# Convert duration to HH:MM:SS\nDURATION_HMS=$(date -u -r \"$DURATION\" +%H:%M:%S 2>\u002Fdev\u002Fnull || echo \"Unknown\")\n```\n\n4. **Check file size** (warn if large for cloud APIs):\n\n```bash\nSIZE_MB=$(du -m \"$AUDIO_FILE\" | cut -f1)\nif [[ $SIZE_MB -gt 25 ]]; then\n    echo \"⚠️  Large file ($FILE_SIZE) - processing may take several minutes\"\nfi\n```\n\n5. **Validate format** (supported: MP3, WAV, M4A, OGG, FLAC, WEBM):\n\n```bash\nEXTENSION=\"${AUDIO_FILE##*.}\"\nSUPPORTED_FORMATS=(\"mp3\" \"wav\" \"m4a\" \"ogg\" \"flac\" \"webm\" \"mp4\")\n\nif [[ ! \" ${SUPPORTED_FORMATS[@]} \" =~ \" ${EXTENSION,,} \" ]]; then\n    echo \"⚠️  Unsupported format: $EXTENSION\"\n    if command -v ffmpeg &>\u002Fdev\u002Fnull; then\n        echo \"? Converting to WAV...\"\n        ffmpeg -i \"$AUDIO_FILE\" -ar 16000 \"${AUDIO_FILE%.*}.wav\" -y\n        AUDIO_FILE=\"${AUDIO_FILE%.*}.wav\"\n    else\n        echo \"❌ Install ffmpeg to convert formats: brew install ffmpeg\"\n        exit 1\n    fi\nfi\n```\n\n\n### Step 3: Generate Markdown Output\n\n**Objective:** Create structured Markdown with metadata, transcription, meeting minutes, and summary.\n\n**Output Template:**\n\n```markdown\n# Audio Transcription Report\n\n## ? Metadata\n\n| Field | Value |\n|-------|-------|\n| **File Name** | {filename} |\n| **File Size** | {file_size} |\n| **Duration** | {duration_hms} |\n| **Language** | {language} ({language_code}) |\n| **Processed Date** | {process_date} |\n| **Speakers Identified** | {num_speakers} |\n| **Transcription Engine** | {engine} (model: {model}) |\n\n\n## ? Meeting Minutes\n\n### Participants\n- {speaker_1}\n- {speaker_2}\n- ...\n\n### Topics Discussed\n1. **{topic_1}** ({timestamp})\n   - {key_point_1}\n   - {key_point_2}\n\n2. **{topic_2}** ({timestamp})\n   - {key_point_1}\n\n### Decisions Made\n- ✅ {decision_1}\n- ✅ {decision_2}\n\n### Action Items\n- [ ] **{action_1}** - Assigned to: {speaker} - Due: {date_if_mentioned}\n- [ ] **{action_2}** - Assigned to: {speaker}\n\n\n*Generated by audio-transcriber skill v1.0.0*  \n*Transcription engine: {engine} | Processing time: {elapsed_time}s*\n```\n\n**Implementation:**\n\nUse Python or bash with AI model (Claude\u002FGPT) for intelligent summarization:\n\n```python\ndef generate_meeting_minutes(segments):\n    \"\"\"Extract topics, decisions, action items from transcription.\"\"\"\n    \n    # Group segments by topic (simple clustering by timestamps)\n    topics = cluster_by_topic(segments)\n    \n    # Identify action items (keywords: \"should\", \"will\", \"need to\", \"action\")\n    action_items = extract_action_items(segments)\n    \n    # Identify decisions (keywords: \"decided\", \"agreed\", \"approved\")\n    decisions = extract_decisions(segments)\n    \n    return {\n        \"topics\": topics,\n        \"decisions\": decisions,\n        \"action_items\": action_items\n    }\n\ndef generate_summary(segments, max_paragraphs=5):\n    \"\"\"Create executive summary using AI (Claude\u002FGPT via API or local model).\"\"\"\n    \n    full_text = \" \".join([s[\"text\"] for s in segments])\n    \n    # Use Chain of Density approach (from prompt-engineer frameworks)\n    summary_prompt = f\"\"\"\n    Summarize the following transcription in {max_paragraphs} concise paragraphs.\n    Focus on key topics, decisions, and action items.\n    \n    Transcription:\n    {full_text}\n    \"\"\"\n    \n    # Call AI model (placeholder - user can integrate Claude API or use local model)\n    summary = call_ai_model(summary_prompt)\n    \n    return summary\n```\n\n**Output file naming:**\n\n```bash\n# v1.1.0: Use timestamp para evitar sobrescrever\nTIMESTAMP=$(date +%Y%m%d-%H%M%S)\nTRANSCRIPT_FILE=\"transcript-${TIMESTAMP}.md\"\nATA_FILE=\"ata-${TIMESTAMP}.md\"\n\necho \"$TRANSCRIPT_CONTENT\" > \"$TRANSCRIPT_FILE\"\necho \"✅ Transcript salvo: $TRANSCRIPT_FILE\"\n\nif [[ -n \"$ATA_CONTENT\" ]]; then\n    echo \"$ATA_CONTENT\" > \"$ATA_FILE\"\n    echo \"✅ Ata salva: $ATA_FILE\"\nfi\n```\n\n\n#### **SCENARIO A: User Provided Custom Prompt**\n\n**Workflow:**\n\n1. **Display user's prompt:**\n   ```\n   ? Prompt fornecido pelo usuário:\n   ┌──────────────────────────────────┐\n   │ [User's prompt preview]          │\n   └──────────────────────────────────┘\n   ```\n\n2. **Automatically improve with prompt-engineer (if available):**\n   ```bash\n   ? Melhorando prompt com prompt-engineer...\n   [Invokes: gh copilot -p \"melhore este prompt: {user_prompt}\"]\n   ```\n\n3. **Show both versions:**\n   ```\n   ✨ Versão melhorada:\n   ┌──────────────────────────────────┐\n   │ Role: Você é um documentador...  │\n   │ Instructions: Transforme...      │\n   │ Steps: 1) ... 2) ...             │\n   │ End Goal: ...                    │\n   └──────────────────────────────────┘\n\n   ? Versão original:\n   ┌──────────────────────────────────┐\n   │ [User's original prompt]         │\n   └──────────────────────────────────┘\n   ```\n\n4. **Ask which to use:**\n   ```bash\n   ? Usar versão melhorada? [s\u002Fn] (default: s):\n   ```\n\n5. **Process with selected prompt:**\n   - If \"s\": use improved\n   - If \"n\": use original\n\n\n#### **LLM Processing (Both Scenarios)**\n\nOnce prompt is finalized:\n\n```python\nfrom rich.progress import Progress, SpinnerColumn, TextColumn\n\ndef process_with_llm(transcript, prompt, cli_tool='claude'):\n    full_prompt = f\"{prompt}\\n\\n---\\n\\nTranscrição:\\n\\n{transcript}\"\n    \n    with Progress(\n        SpinnerColumn(),\n        TextColumn(\"[progress.description]{task.description}\"),\n        transient=True\n    ) as progress:\n        progress.add_task(\n            description=f\"? Processando com {cli_tool}...\",\n            total=None\n        )\n        \n        if cli_tool == 'claude':\n            result = subprocess.run(\n                ['claude', '-'],\n                input=full_prompt,\n                capture_output=True,\n                text=True,\n                timeout=300  # 5 minutes\n            )\n        elif cli_tool == 'gh-copilot':\n            result = subprocess.run(\n                ['gh', 'copilot', 'suggest', '-t', 'shell', full_prompt],\n                capture_output=True,\n                text=True,\n                timeout=300\n            )\n    \n    if result.returncode == 0:\n        return result.stdout.strip()\n    else:\n        return None\n```\n\n**Progress output:**\n```\n? Processando com claude... ⠋\n[After completion:]\n✅ Ata gerada com sucesso!\n```\n\n\n#### **Final Output**\n\n**Success (both files):**\n```bash\n? Salvando arquivos...\n\n✅ Arquivos criados:\n  - transcript-20260203-023045.md  (transcript puro)\n  - ata-20260203-023045.md         (processado com LLM)\n\n? Removidos arquivos temporários: metadata.json, transcription.json\n\n✅ Concluído! Tempo total: 3m 45s\n```\n\n**Transcript only (user declined LLM):**\n```bash\n? Salvando arquivos...\n\n✅ Arquivo criado:\n  - transcript-20260203-023045.md\n\nℹ️  Ata não gerada (processamento LLM recusado pelo usuário)\n\n? Removidos arquivos temporários: metadata.json, transcription.json\n\n✅ Concluído!\n```\n\n\n### Step 5: Display Results Summary\n\n**Objective:** Show completion status and next steps.\n\n**Output:**\n\n```bash\necho \"\"\necho \"✅ Transcription Complete!\"\necho \"\"\necho \"? Results:\"\necho \"  File: $OUTPUT_FILE\"\necho \"  Language: $LANGUAGE\"\necho \"  Duration: $DURATION_HMS\"\necho \"  Speakers: $NUM_SPEAKERS\"\necho \"  Words: $WORD_COUNT\"\necho \"  Processing time: ${ELAPSED_TIME}s\"\necho \"\"\necho \"? Generated:\"\necho \"  - $OUTPUT_FILE (Markdown report)\"\n[if alternative formats:]\necho \"  - ${OUTPUT_FILE%.*}.srt (Subtitles)\"\necho \"  - ${OUTPUT_FILE%.*}.json (Structured data)\"\necho \"\"\necho \"? Next steps:\"\necho \"  1. Review meeting minutes and action items\"\necho \"  2. Share report with participants\"\necho \"  3. Track action items to completion\"\n```\n\n\n## Example Usage\n\n### **Example 1: Basic Transcription**\n\n**User Input:**\n```bash\ncopilot> transcribe audio to markdown: meeting-2026-02-02.mp3\n```\n\n**Skill Output:**\n\n```bash\n✅ Faster-Whisper detected (optimized)\n✅ ffmpeg available (format conversion enabled)\n\n? File: meeting-2026-02-02.mp3\n? Size: 12.3 MB\n⏱️  Duration: 00:45:32\n\n?️  Processing...\n[████████████████████] 100%\n\n✅ Language detected: Portuguese (pt-BR)\n? Speakers identified: 4\n? Generating Markdown output...\n\n✅ Transcription Complete!\n\n? Results:\n  File: meeting-2026-02-02.md\n  Language: pt-BR\n  Duration: 00:45:32\n  Speakers: 4\n  Words: 6,842\n  Processing time: 127s\n\n? Generated:\n  - meeting-2026-02-02.md (Markdown report)\n\n? Next steps:\n  1. Review meeting minutes and action items\n  2. Share report with participants\n  3. Track action items to completion\n```\n\n\n### **Example 3: Batch Processing**\n\n**User Input:**\n```bash\ncopilot> transcreva estes áudios: recordings\u002F*.mp3\n```\n\n**Skill Output:**\n\n```bash\n? Batch mode: 5 files found\n  1. team-standup.mp3\n  2. client-call.mp3\n  3. brainstorm-session.mp3\n  4. product-demo.mp3\n  5. retrospective.mp3\n\n?️  Processing batch...\n\n[1\u002F5] team-standup.mp3 ✅ (2m 34s)\n[2\u002F5] client-call.mp3 ✅ (15m 12s)\n[3\u002F5] brainstorm-session.mp3 ✅ (8m 47s)\n[4\u002F5] product-demo.mp3 ✅ (22m 03s)\n[5\u002F5] retrospective.mp3 ✅ (11m 28s)\n\n✅ Batch Complete!\n? Generated 5 Markdown reports\n⏱️  Total processing time: 6m 15s\n```\n\n\n### **Example 5: Large File Warning**\n\n**User Input:**\n```bash\ncopilot> transcribe audio to markdown: conference-keynote.mp3\n```\n\n**Skill Output:**\n\n```bash\n✅ Faster-Whisper detected (optimized)\n\n? File: conference-keynote.mp3\n? Size: 87.2 MB\n⏱️  Duration: 02:15:47\n⚠️  Large file (87.2 MB) - processing may take several minutes\n\nContinue? [Y\u002Fn]:\n```\n\n**User:** `Y`\n\n```bash\n?️  Processing... (this may take 10-15 minutes)\n[████░░░░░░░░░░░░░░░░] 20% - Estimated time remaining: 12m\n```\n\n\nThis skill is **platform-agnostic** and works in any terminal context where GitHub Copilot CLI is available. It does not depend on specific project configurations or external APIs, following the zero-configuration philosophy.\n","## Purpose\n\nThis skill automates audio-to-text transcription with professional Markdown output, extracting rich technical metadata (speakers, timestamps, language, file size, duration) and generating structured meeting minutes and executive summaries. It uses Faster-Whisper or Whisper with zero configuration, working universally across projects without hardcoded paths or API keys.\n\nInspired by tools like Plaud, this skill transforms raw audio recordings into actionable documentation, making it ideal for meetings, interviews, lectures, and content analysis.\n\n## When to Use\nInvoke this skill when:\n\n- User needs to transcribe audio\u002Fvideo files to text\n- User wants meeting minutes automatically generated from recordings\n- User requires speaker identification (diarization) in conversations\n- User needs subtitles\u002Fcaptions (SRT, VTT formats)\n- User wants executive summaries of long audio content\n- User asks variations of \"transcribe this audio\", \"convert audio to text\", \"generate meeting notes from recording\"\n- User has audio files in common formats (MP3, WAV, M4A, OGG, FLAC, WEBM)\n\n## Workflow\n\n### Step 0: Discovery (Auto-detect Transcription Tools)\n\n**Objective:** Identify available transcription engines without user configuration.\n\n**Actions:**\n\nRun detection commands to find installed tools:\n\n```bash\n# Check for Faster-Whisper (preferred - 4-5x faster)\nif python3 -c \"import faster_whisper\" 2>\u002Fdev\u002Fnull; then\n    TRANSCRIBER=\"faster-whisper\"\n    echo \"✅ Faster-Whisper detected (optimized)\"\n# Fallback to original Whisper\nelif python3 -c \"import whisper\" 2>\u002Fdev\u002Fnull; then\n    TRANSCRIBER=\"whisper\"\n    echo \"✅ OpenAI Whisper detected\"\nelse\n    TRANSCRIBER=\"none\"\n    echo \"⚠️  No transcription tool found\"\nfi\n\n# Check for ffmpeg (audio format conversion)\nif command -v ffmpeg &>\u002Fdev\u002Fnull; then\n    echo \"✅ ffmpeg available (format conversion enabled)\"\nelse\n    echo \"ℹ️  ffmpeg not found (limited format support)\"\nfi\n```\n\n**If no transcriber found:**\n\nOffer automatic installation using the provided script:\n\n```bash\necho \"⚠️  No transcription tool found\"\necho \"\"\necho \"? Auto-install dependencies? (Recommended)\"\nread -p \"Run installation script? [Y\u002Fn]: \" AUTO_INSTALL\n\nif [[ ! \"$AUTO_INSTALL\" =~ ^[Nn] ]]; then\n    # Get skill directory (works for both repo and symlinked installations)\n    SKILL_DIR=\"$(cd \"$(dirname \"${BASH_SOURCE[0]}\")\" && pwd)\"\n    \n    # Run installation script\n    if [[ -f \"$SKILL_DIR\u002Fscripts\u002Finstall-requirements.sh\" ]]; then\n        bash \"$SKILL_DIR\u002Fscripts\u002Finstall-requirements.sh\"\n    else\n        echo \"❌ Installation script not found\"\n        echo \"\"\n        echo \"? Manual installation:\"\n        echo \"  pip install faster-whisper  # Recommended\"\n        echo \"  pip install openai-whisper  # Alternative\"\n        echo \"  brew install ffmpeg         # Optional (macOS)\"\n        exit 1\n    fi\n    \n    # Verify installation succeeded\n    if python3 -c \"import faster_whisper\" 2>\u002Fdev\u002Fnull || python3 -c \"import whisper\" 2>\u002Fdev\u002Fnull; then\n        echo \"✅ Installation successful! Proceeding with transcription...\"\n    else\n        echo \"❌ Installation failed. Please install manually.\"\n        exit 1\n    fi\nelse\n    echo \"\"\n    echo \"? Manual installation required:\"\n    echo \"\"\n    echo \"Recommended (fastest):\"\n    echo \"  pip install faster-whisper\"\n    echo \"\"\n    echo \"Alternative (original):\"\n    echo \"  pip install openai-whisper\"\n    echo \"\"\n    echo \"Optional (format conversion):\"\n    echo \"  brew install ffmpeg  # macOS\"\n    echo \"  apt install ffmpeg   # Linux\"\n    echo \"\"\n    exit 1\nfi\n```\n\nThis ensures users can install dependencies with one confirmation, or opt for manual installation if preferred.\n\n**If transcriber found:**\n\nProceed to Step 0b (CLI Detection).\n\n\n### Step 1: Validate Audio File\n\n**Objective:** Verify file exists, check format, and extract metadata.\n\n**Actions:**\n\n1. **Accept file path or URL** from user:\n   - Local file: `meeting.mp3`\n   - URL: `https:\u002F\u002Fexample.com\u002Faudio.mp3` (download to temp directory)\n\n2. **Verify file exists:**\n\n```bash\nif [[ ! -f \"$AUDIO_FILE\" ]]; then\n    echo \"❌ File not found: $AUDIO_FILE\"\n    exit 1\nfi\n```\n\n3. **Extract metadata** using ffprobe or file utilities:\n\n```bash\n# Get file size\nFILE_SIZE=$(du -h \"$AUDIO_FILE\" | cut -f1)\n\n# Get duration and format using ffprobe\nDURATION=$(ffprobe -v error -show_entries format=duration \\\n    -of default=noprint_wrappers=1:nokey=1 \"$AUDIO_FILE\" 2>\u002Fdev\u002Fnull)\nFORMAT=$(ffprobe -v error -select_streams a:0 -show_entries \\\n    stream=codec_name -of default=noprint_wrappers=1:nokey=1 \"$AUDIO_FILE\" 2>\u002Fdev\u002Fnull)\n\n# Convert duration to HH:MM:SS\nDURATION_HMS=$(date -u -r \"$DURATION\" +%H:%M:%S 2>\u002Fdev\u002Fnull || echo \"Unknown\")\n```\n\n4. **Check file size** (warn if large for cloud APIs):\n\n```bash\nSIZE_MB=$(du -m \"$AUDIO_FILE\" | cut -f1)\nif [[ $SIZE_MB -gt 25 ]]; then\n    echo \"⚠️  Large file ($FILE_SIZE) - processing may take several minutes\"\nfi\n```\n\n5. **Validate format** (supported: MP3, WAV, M4A, OGG, FLAC, WEBM):\n\n```bash\nEXTENSION=\"${AUDIO_FILE##*.}\"\nSUPPORTED_FORMATS=(\"mp3\" \"wav\" \"m4a\" \"ogg\" \"flac\" \"webm\" \"mp4\")\n\nif [[ ! \" ${SUPPORTED_FORMATS[@]} \" =~ \" ${EXTENSION,,} \" ]]; then\n    echo \"⚠️  Unsupported format: $EXTENSION\"\n    if command -v ffmpeg &>\u002Fdev\u002Fnull; then\n        echo \"? Converting to WAV...\"\n        ffmpeg -i \"$AUDIO_FILE\" -ar 16000 \"${AUDIO_FILE%.*}.wav\" -y\n        AUDIO_FILE=\"${AUDIO_FILE%.*}.wav\"\n    else\n        echo \"❌ Install ffmpeg to convert formats: brew install ffmpeg\"\n        exit 1\n    fi\nfi\n```\n\n\n### Step 3: Generate Markdown Output\n\n**Objective:** Create structured Markdown with metadata, transcription, meeting minutes, and summary.\n\n**Output Template:**\n\n```markdown\n# Audio Transcription Report\n\n## ? Metadata\n\n| Field | Value |\n|-------|-------|\n| **File Name** | {filename} |\n| **File Size** | {file_size} |\n| **Duration** | {duration_hms} |\n| **Language** | {language} ({language_code}) |\n| **Processed Date** | {process_date} |\n| **Speakers Identified** | {num_speakers} |\n| **Transcription Engine** | {engine} (model: {model}) |\n\n\n## ? Meeting Minutes\n\n### Participants\n- {speaker_1}\n- {speaker_2}\n- ...\n\n### Topics Discussed\n1. **{topic_1}** ({timestamp})\n   - {key_point_1}\n   - {key_point_2}\n\n2. **{topic_2}** ({timestamp})\n   - {key_point_1}\n\n### Decisions Made\n- ✅ {decision_1}\n- ✅ {decision_2}\n\n### Action Items\n- [ ] **{action_1}** - Assigned to: {speaker} - Due: {date_if_mentioned}\n- [ ] **{action_2}** - Assigned to: {speaker}\n\n\n*Generated by audio-transcriber skill v1.0.0*  \n*Transcription engine: {engine} | Processing time: {elapsed_time}s*\n```\n\n**Implementation:**\n\nUse Python or bash with AI model (Claude\u002FGPT) for intelligent summarization:\n\n```python\ndef generate_meeting_minutes(segments):\n    \"\"\"Extract topics, decisions, action items from transcription.\"\"\"\n    \n    # Group segments by topic (simple clustering by timestamps)\n    topics = cluster_by_topic(segments)\n    \n    # Identify action items (keywords: \"should\", \"will\", \"need to\", \"action\")\n    action_items = extract_action_items(segments)\n    \n    # Identify decisions (keywords: \"decided\", \"agreed\", \"approved\")\n    decisions = extract_decisions(segments)\n    \n    return {\n        \"topics\": topics,\n        \"decisions\": decisions,\n        \"action_items\": action_items\n    }\n\ndef generate_summary(segments, max_paragraphs=5):\n    \"\"\"Create executive summary using AI (Claude\u002FGPT via API or local model).\"\"\"\n    \n    full_text = \" \".join([s[\"text\"] for s in segments])\n    \n    # Use Chain of Density approach (from prompt-engineer frameworks)\n    summary_prompt = f\"\"\"\n    Summarize the following transcription in {max_paragraphs} concise paragraphs.\n    Focus on key topics, decisions, and action items.\n    \n    Transcription:\n    {full_text}\n    \"\"\"\n    \n    # Call AI model (placeholder - user can integrate Claude API or use local model)\n    summary = call_ai_model(summary_prompt)\n    \n    return summary\n```\n\n**Output file naming:**\n\n```bash\n# v1.1.0: Use timestamp para evitar sobrescrever\nTIMESTAMP=$(date +%Y%m%d-%H%M%S)\nTRANSCRIPT_FILE=\"transcript-${TIMESTAMP}.md\"\nATA_FILE=\"ata-${TIMESTAMP}.md\"\n\necho \"$TRANSCRIPT_CONTENT\" > \"$TRANSCRIPT_FILE\"\necho \"✅ Transcript salvo: $TRANSCRIPT_FILE\"\n\nif [[ -n \"$ATA_CONTENT\" ]]; then\n    echo \"$ATA_CONTENT\" > \"$ATA_FILE\"\n    echo \"✅ Ata salva: $ATA_FILE\"\nfi\n```\n\n\n#### **SCENARIO A: User Provided Custom Prompt**\n\n**Workflow:**\n\n1. **Display user's prompt:**\n   ```\n   ? Prompt fornecido pelo usuário:\n   ┌──────────────────────────────────┐\n   │ [User's prompt preview]          │\n   └──────────────────────────────────┘\n   ```\n\n2. **Automatically improve with prompt-engineer (if available):**\n   ```bash\n   ? Melhorando prompt com prompt-engineer...\n   [Invokes: gh copilot -p \"melhore este prompt: {user_prompt}\"]\n   ```\n\n3. **Show both versions:**\n   ```\n   ✨ Versão melhorada:\n   ┌──────────────────────────────────┐\n   │ Role: Você é um documentador...  │\n   │ Instructions: Transforme...      │\n   │ Steps: 1) ... 2) ...             │\n   │ End Goal: ...                    │\n   └──────────────────────────────────┘\n\n   ? Versão original:\n   ┌──────────────────────────────────┐\n   │ [User's original prompt]         │\n   └──────────────────────────────────┘\n   ```\n\n4. **Ask which to use:**\n   ```bash\n   ? Usar versão melhorada? [s\u002Fn] (default: s):\n   ```\n\n5. **Process with selected prompt:**\n   - If \"s\": use improved\n   - If \"n\": use original\n\n\n#### **LLM Processing (Both Scenarios)**\n\nOnce prompt is finalized:\n\n```python\nfrom rich.progress import Progress, SpinnerColumn, TextColumn\n\ndef process_with_llm(transcript, prompt, cli_tool='claude'):\n    full_prompt = f\"{prompt}\\n\\n---\\n\\nTranscrição:\\n\\n{transcript}\"\n    \n    with Progress(\n        SpinnerColumn(),\n        TextColumn(\"[progress.description]{task.description}\"),\n        transient=True\n    ) as progress:\n        progress.add_task(\n            description=f\"? Processando com {cli_tool}...\",\n            total=None\n        )\n        \n        if cli_tool == 'claude':\n            result = subprocess.run(\n                ['claude', '-'],\n                input=full_prompt,\n                capture_output=True,\n                text=True,\n                timeout=300  # 5 minutes\n            )\n        elif cli_tool == 'gh-copilot':\n            result = subprocess.run(\n                ['gh', 'copilot', 'suggest', '-t', 'shell', full_prompt],\n                capture_output=True,\n                text=True,\n                timeout=300\n            )\n    \n    if result.returncode == 0:\n        return result.stdout.strip()\n    else:\n        return None\n```\n\n**Progress output:**\n```\n? Processando com claude... ⠋\n[After completion:]\n✅ Ata gerada com sucesso!\n```\n\n\n#### **Final Output**\n\n**Success (both files):**\n```bash\n? Salvando arquivos...\n\n✅ Arquivos criados:\n  - transcript-20260203-023045.md  (transcript puro)\n  - ata-20260203-023045.md         (processado com LLM)\n\n? Removidos arquivos temporários: metadata.json, transcription.json\n\n✅ Concluído! Tempo total: 3m 45s\n```\n\n**Transcript only (user declined LLM):**\n```bash\n? Salvando arquivos...\n\n✅ Arquivo criado:\n  - transcript-20260203-023045.md\n\nℹ️  Ata não gerada (processamento LLM recusado pelo usuário)\n\n? Removidos arquivos temporários: metadata.json, transcription.json\n\n✅ Concluído!\n```\n\n\n### Step 5: Display Results Summary\n\n**Objective:** Show completion status and next steps.\n\n**Output:**\n\n```bash\necho \"\"\necho \"✅ Transcription Complete!\"\necho \"\"\necho \"? Results:\"\necho \"  File: $OUTPUT_FILE\"\necho \"  Language: $LANGUAGE\"\necho \"  Duration: $DURATION_HMS\"\necho \"  Speakers: $NUM_SPEAKERS\"\necho \"  Words: $WORD_COUNT\"\necho \"  Processing time: ${ELAPSED_TIME}s\"\necho \"\"\necho \"? Generated:\"\necho \"  - $OUTPUT_FILE (Markdown report)\"\n[if alternative formats:]\necho \"  - ${OUTPUT_FILE%.*}.srt (Subtitles)\"\necho \"  - ${OUTPUT_FILE%.*}.json (Structured data)\"\necho \"\"\necho \"? Next steps:\"\necho \"  1. Review meeting minutes and action items\"\necho \"  2. Share report with participants\"\necho \"  3. Track action items to completion\"\n```\n\n\n## Example Usage\n\n### **Example 1: Basic Transcription**\n\n**User Input:**\n```bash\ncopilot> transcribe audio to markdown: meeting-2026-02-02.mp3\n```\n\n**Skill Output:**\n\n```bash\n✅ Faster-Whisper detected (optimized)\n✅ ffmpeg available (format conversion enabled)\n\n? File: meeting-2026-02-02.mp3\n? Size: 12.3 MB\n⏱️  Duration: 00:45:32\n\n?️  Processing...\n[████████████████████] 100%\n\n✅ Language detected: Portuguese (pt-BR)\n? Speakers identified: 4\n? Generating Markdown output...\n\n✅ Transcription Complete!\n\n? Results:\n  File: meeting-2026-02-02.md\n  Language: pt-BR\n  Duration: 00:45:32\n  Speakers: 4\n  Words: 6,842\n  Processing time: 127s\n\n? Generated:\n  - meeting-2026-02-02.md (Markdown report)\n\n? Next steps:\n  1. Review meeting minutes and action items\n  2. Share report with participants\n  3. Track action items to completion\n```\n\n\n### **Example 3: Batch Processing**\n\n**User Input:**\n```bash\ncopilot> transcreva estes áudios: recordings\u002F*.mp3\n```\n\n**Skill Output:**\n\n```bash\n? Batch mode: 5 files found\n  1. team-standup.mp3\n  2. client-call.mp3\n  3. brainstorm-session.mp3\n  4. product-demo.mp3\n  5. retrospective.mp3\n\n?️  Processing batch...\n\n[1\u002F5] team-standup.mp3 ✅ (2m 34s)\n[2\u002F5] client-call.mp3 ✅ (15m 12s)\n[3\u002F5] brainstorm-session.mp3 ✅ (8m 47s)\n[4\u002F5] product-demo.mp3 ✅ (22m 03s)\n[5\u002F5] retrospective.mp3 ✅ (11m 28s)\n\n✅ Batch Complete!\n? Generated 5 Markdown reports\n⏱️  Total processing time: 6m 15s\n```\n\n\n### **Example 5: Large File Warning**\n\n**User Input:**\n```bash\ncopilot> transcribe audio to markdown: conference-keynote.mp3\n```\n\n**Skill Output:**\n\n```bash\n✅ Faster-Whisper detected (optimized)\n\n? File: conference-keynote.mp3\n? Size: 87.2 MB\n⏱️  Duration: 02:15:47\n⚠️  Large file (87.2 MB) - processing may take several minutes\n\nContinue? [Y\u002Fn]:\n```\n\n**User:** `Y`\n\n```bash\n?️  Processing... (this may take 10-15 minutes)\n[████░░░░░░░░░░░░░░░░] 20% - Estimated time remaining: 12m\n```\n\n\nThis skill is **platform-agnostic** and works in any terminal context where GitHub Copilot CLI is available. It does not depend on specific project configurations or external APIs, following the zero-configuration philosophy.\n",{"name":5,"risk":38,"tags":52,"source":39,"category":53,"date_added":54,"description":7},"[audio, transcription, whisper, meeting-minutes, speech-to-text]","content","2026-02-27","main","c173cdffe08e788a99b91928a831b476b8434838","ecosystem\u002Fskills\u002Faudio_transcriber",0,"https:\u002F\u002Fgithub.com\u002FLongLeo287\u002FOmniClaw","https:\u002F\u002Fportfolio-longleo.vercel.app\u002Flabs\u002Fomniclaw.html","https:\u002F\u002Fgithub.com\u002FLongLeo287","https:\u002F\u002Fwww.facebook.com\u002FResourceRookie2023","2026-11-14 08:48:28",{"source":8,"skillName":5,"commands":65,"prompt":75,"zipAvailable":76},[66,69,72],{"tool":67,"command":68},"npx","npx skills add LongLeo287\u002FOmniClaw --skill audio_transcriber",{"tool":70,"command":71},"bunx","bunx skills add LongLeo287\u002FOmniClaw --skill audio_transcriber",{"tool":73,"command":74},"pnpm","pnpm dlx skills add LongLeo287\u002FOmniClaw --skill audio_transcriber","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FLongLeo287\u002FOmniClaw\u002Fblob\u002Fmain\u002Fecosystem\u002Fskills\u002Faudio_transcriber\u002FSKILL.md",true,[78,84,88,92],{"path":79,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":82,"sourceHtmlUrl":83},"_DIR_IDENTITY.md","REFERENCE","参考文档",156,"https:\u002F\u002Fgithub.com\u002FLongLeo287\u002FOmniClaw\u002Fblob\u002Fmain\u002Fecosystem\u002Fskills\u002Faudio_transcriber\u002F_DIR_IDENTITY.md",{"path":85,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":86,"sourceHtmlUrl":87},"changelog_1.md",5462,"https:\u002F\u002Fgithub.com\u002FLongLeo287\u002FOmniClaw\u002Fblob\u002Fmain\u002Fecosystem\u002Fskills\u002Faudio_transcriber\u002Fchangelog_1.md",{"path":89,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":90,"sourceHtmlUrl":91},"README.md",9530,"https:\u002F\u002Fgithub.com\u002FLongLeo287\u002FOmniClaw\u002Fblob\u002Fmain\u002Fecosystem\u002Fskills\u002Faudio_transcriber\u002FREADME.md",{"path":93,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":94,"sourceHtmlUrl":95},"schema.json",453,"https:\u002F\u002Fgithub.com\u002FLongLeo287\u002FOmniClaw\u002Fblob\u002Fmain\u002Fecosystem\u002Fskills\u002Faudio_transcriber\u002Fschema.json",[97,136,157,185,201,224],{"skillNo":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":101,"categories":102,"keywords":115,"publishedAt":47},"01M6JR0KQP4MHT32TN40D9CBFX","UI-UX Pro Max","Advanced UI\u002FUX analysis and generation skills with data-driven patterns. Python scripts (core.py, search.py) for systematic UI pattern extraction, design system generation, and component search from curated data sources.","https:\u002F\u002Fgithub.com\u002FLongLeo287\u002FOmniClaw\u002Fblob\u002Fmain\u002Fecosystem\u002Fui_components\u002Fui_ux_pro_max\u002FSKILL.md",[103,106,107,112],{"categoryNo":104,"categoryName":105,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000015","视觉设计",{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":108,"categoryName":109,"taxonomyType":110,"taxonomyTypeDesc":111},"01JBSEEDCAT000000000000032","设计师","PROFESSION","适用职业",{"categoryNo":113,"categoryName":114,"taxonomyType":110,"taxonomyTypeDesc":111},"01JBSEEDCAT000000000000033","数据分析师",[116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135],"ux","ui","ui ux pro max","max","pro","ui ux","design","component","search","analysis","team","rd","ui-ux","active","systematic","scripts","pattern extraction","sources","advanced","data-driven",{"skillNo":137,"name":138,"nameZh":6,"description":139,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":140,"categories":141,"keywords":143,"publishedAt":47},"01M6JR0EF5AEQHP2FSTB57H05P","shadcn-ui","Expert guidance for integrating and building applications with shadcn\u002Fui components, including component discovery, installation, customization, and best practices.","https:\u002F\u002Fgithub.com\u002FLongLeo287\u002FOmniClaw\u002Fblob\u002Fmain\u002Fecosystem\u002Fui_components\u002Fshadcn_ui\u002FSKILL.md",[142],{"categoryNo":108,"categoryName":109,"taxonomyType":110,"taxonomyTypeDesc":111},[144,117,145,146,123,147,148,149,150,126,151,152,153,129,154,155,33,156],"shadcn","shadcn ui","components","core","best practices","applications","discovery","customization","guidance","expert","installation","ui components","for",{"skillNo":158,"name":159,"nameZh":6,"description":160,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":161,"categories":162,"keywords":166,"publishedAt":47},"01M6JR08F51BEN17RHFYX65WVE","zustand_store_ts","Create Zustand stores with TypeScript, subscribeWithSelector middleware, and proper state\u002Faction separation. Use when building React state management, creating global stores, or implementing reactive state patterns with Zustand.","https:\u002F\u002Fgithub.com\u002FLongLeo287\u002FOmniClaw\u002Fblob\u002Fmain\u002Fecosystem\u002Fskills\u002Fzustand_store_ts\u002FSKILL.md",[163],{"categoryNo":164,"categoryName":165,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000009","前端开发",[167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,33,183,184],"zustand store ts","zustand","store","typescript","microsoft","react","create","state management","state","react state management","management","global","use","patterns","separation","middleware","action","implementing",{"skillNo":186,"name":168,"nameZh":6,"description":187,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":188,"categories":189,"keywords":190,"publishedAt":47},"01M6JR00TDMVAFFCZXZ04AN7Q7","Zustand state management guide. Use when working with store code (src\u002Fstore\u002F**), implementing actions, managing state, or creating slices. Triggers on Zustand store development, state management questions, or action implementation.","https:\u002F\u002Fgithub.com\u002FLongLeo287\u002FOmniClaw\u002Fblob\u002Fmain\u002Fecosystem\u002Fskills\u002Fzustand\u002FSKILL.md",[],[168,191,174,192,175,193,194,177,195,179,196,197,198,169,33,199,200,183,184],"development","code","actions","src","guide","implementation","zustand state","triggers","questions","zustand state management",{"skillNo":202,"name":203,"nameZh":6,"description":204,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":205,"categories":206,"keywords":209,"publishedAt":47},"01M6JQZSWV4KWJP112G6XHEP44","youtube_summarizer","Automatically fetch YouTube video transcripts, generate structured summaries, and send full transcripts to messaging platforms. Detects YouTube URLs and provides metadata, key insights, and downloadable transcripts.","https:\u002F\u002Fgithub.com\u002FLongLeo287\u002FOmniClaw\u002Fblob\u002Fmain\u002Fecosystem\u002Fskills\u002Fyoutube_summarizer\u002FSKILL.md",[207,208],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},[210,211,212,213,214,215,216,217,218,29,219,220,221,222,223,30],"youtube","youtube summarizer","summarizer","generate","metadata","messaging","Video","fetch","telegram","summarization","structured","full","insights","transcripts",{"skillNo":225,"name":226,"nameZh":6,"description":227,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":228,"categories":229,"keywords":230,"publishedAt":47},"01M6JQZN3QJB6B77JGKJD62DQT","youtube_clipper","YouTube 视频智能剪辑工具。下载视频和字幕，AI 分析生成精细章节（几分钟级别）， 用户选择片段后自动剪辑、翻译字幕为中英双语、烧录字幕到视频，并生成总结文案。 使用场景：当用户需要剪辑 YouTube 视频、生成短视频片段、制作双语字幕版本时。 关键词：视频剪辑、YouTube、字幕翻译、双语字幕、视频下载、clip video","https:\u002F\u002Fgithub.com\u002FLongLeo287\u002FOmniClaw\u002Fblob\u002Fmain\u002Fecosystem\u002Fskills\u002Fyoutube_clipper\u002FSKILL.md",[],[210,231,232,216,233,234,235,236,237,238,239,240,241,242,243,244,245],"clipper","youtube clipper","Bash","下载","视频","生成","关键词","翻译","read","write","clip","总结","工具","文案","sonnet"]