[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6ECD2D6X4MSBC3TKVSBETJ9":3,"skill-same:01M6ECD2D6X4MSBC3TKVSBETJ9":120},{"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":28,"publishedAt":49,"linkOnly":50,"markdownContent":51,"markdownBody":52,"frontMatter":53,"refName":65,"commitSha":66,"skillDirectory":67,"forksCount":68,"repositoryHtmlUrl":69,"repositoryHomepageUrl":6,"ownerHtmlUrl":70,"ownerBlogUrl":71,"ownerDescription":72,"updatedAt":73,"install":74,"resources":87},"01M6ECD2D6X4MSBC3TKVSBETJ9","assemblyai-streaming",null,"This skill should be used when working with AssemblyAI’s Speech-to-Text and LLM Gateway APIs, especially for streaming\u002Flive transcription, meeting notetakers, and voice agents that need low-latency transcripts and audio analysis.","levalencia\u002Fagent-god-mode","levalencia","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F6962857?v=4","01M6EB0FZZ4BPQPC4HET4QY6NM","Luis Valencia ","01M6EB0G4HWHPWW0H9X7QKSAHP",3,"MIT","https:\u002F\u002Fgithub.com\u002Flevalencia\u002Fagent-god-mode\u002Fblob\u002Fmain\u002Forganized-skills\u002Fassembly-ai-streaming\u002FSKILL.md",[18,23],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",{"categoryNo":24,"categoryName":25,"taxonomyType":26,"taxonomyTypeDesc":27},"01JBSEEDCAT000000000000034","行政与办公人员","PROFESSION","适用职业",[29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48],"streaming","assemblyai streaming","grep","live","latency","speech-to-text","transcription","edit","write","read","llm gateway","python","text","voice","Bash","Audio","analysis","https","gateway","speech","2026-11-10 10:12:30",false,"---\nname: assemblyai-streaming\ndescription: This skill should be used when working with AssemblyAI’s Speech-to-Text and LLM Gateway APIs, especially for streaming\u002Flive transcription, meeting notetakers, and voice agents that need low-latency transcripts and audio analysis.\nlicense: MIT\nallowed-tools:\n  - Read\n  - Write\n  - Edit\n  - Grep\n  - Glob\n  - Bash\n  - Python\nmetadata:\n  skill-version: \"1.0.0\"\n  upstream-docs: \"https:\u002F\u002Fwww.assemblyai.com\u002Fdocs\"\n  focus: \"streaming-stt, meeting-notetaker, voice-agent, llm-gateway\"\n---\n\n# AssemblyAI Streaming & Live Transcription Skill\n\n## Overview\n\nUse this skill to build and maintain code that talks to AssemblyAI’s:\n\n- **Streaming Speech-to-Text (STT)** via WebSockets (`wss:\u002F\u002Fstreaming.assemblyai.com\u002Fv3\u002Fws`)\n- **Async \u002F pre-recorded STT** via REST (`https:\u002F\u002Fapi.assemblyai.com\u002Fv2\u002Ftranscript`)\n- **LLM Gateway** for applying Claude\u002FGPT\u002FGemini-style models to transcripts (`https:\u002F\u002Fllm-gateway.assemblyai.com`)\n\nThe emphasis is on **streaming\u002Flive transcription**, **meeting notetakers**, and **voice agents**, while still covering async workflows and post-processing.\n\nThis skill assumes a Claude Code environment with access to Python (preferred) and Bash.\n\n---\n\n## When to Use\n\nUse this skill when:\n\n- Implementing **real-time transcription** from a microphone, telephony stream, or audio file.\n- Building a **live meeting notetaker** (Zoom\u002FTeams\u002FMeet), especially with summaries, action items, and highlights.\n- Implementing a **voice agent** where latency and natural turn-taking matter.\n- Migrating from other STT providers (OpenAI\u002FDeepgram\u002FGoogle\u002FAWS\u002Fetc.) to AssemblyAI.\n- Applying **LLMs to audio** via LLM Gateway for summaries, Q&A, topic tagging, or custom prompts.\n\nDo **not** use this skill when:\n\n- The task is generic HTTP client usage with no AssemblyAI-specific logic.\n- The request clearly targets a different STT vendor.\n- The environment cannot safely store or use an API key.\n\n---\n\n## AssemblyAI Mental Model\n\n### 1. Products to care about\n\n1. **Pre-recorded Speech-to-Text (Async)**\n   - REST API: `POST \u002Fv2\u002Ftranscript` → `GET \u002Fv2\u002Ftranscript\u002F{id}`\n   - Designed for files from URLs, uploads, S3, etc.\n   - Supports extra models: summarization, topic detection, sentiment, PII redaction, chapters, etc.\n\n2. **Streaming Speech-to-Text**\n   - WebSocket: `wss:\u002F\u002Fstreaming.assemblyai.com\u002Fv3\u002Fws`\n   - Low-latency, immutable transcripts (~300ms).\n   - Turn detection built in; fits voice agents and live captioning.\n\n3. **LLM Gateway**\n   - REST API: `POST \u002Fv1\u002Fchat\u002Fcompletions` at `https:\u002F\u002Fllm-gateway.assemblyai.com`\n   - Unified access to multiple LLMs (Claude, GPT, Gemini, etc.).\n   - Designed for “LLM over transcripts” workflows.\n\n### 2. Key model knobs (Async)\n\n- `speech_models`: `[\"slam-1\", \"universal\"]` etc.\n  - **Slam-1**: best English accuracy + keyterms_prompt, good for medical\u002Ftechnical conversations.\n  - **Universal**: multilingual coverage; good default if language is unknown.\n- `language_code` vs `language_detection`:\n  - Use `language_code` when the language is known.\n  - Use `language_detection: true` when unknown; optionally set `language_confidence_threshold`.\n- `keyterms_prompt`:\n  - Domain words\u002Fphrases to boost (med terms, product names, etc.).\n- Extra intelligence: `summarization`, `iab_categories`, `content_safety`, `entity_detection`, `auto_chapters`, `sentiment_analysis`, `speaker_labels`, `auto_highlights`, `redact_pii`, etc.\n\n### 3. Key model knobs (Streaming)\n\nConnection URL:\n\n- US: `wss:\u002F\u002Fstreaming.assemblyai.com\u002Fv3\u002Fws`\n- EU: `wss:\u002F\u002Fstreaming.eu.assemblyai.com\u002Fv3\u002Fws`\n\nImportant query parameters:\n\n- `sample_rate` (required): e.g. `16000`\n- `format_turns` (bool): return formatted final transcripts; **avoid** for low-latency voice agents.\n- `speech_model`: `universal-streaming-english` (default) or `universal-streaming-multi`.\n- `keyterms_p\n\nrompt`: JSON-encoded list of terms, e.g. `[\"AssemblyAI\", \"Slam-1\", \"Keanu Reeves\"]`.\n- Turn detection:\n  - `end_of_turn_confidence_threshold` (0.0–1.0, default ~0.4)\n  - `min_end_of_turn_silence_when_confident` (ms, default ~400)\n  - `max_turn_silence` (ms, default ~1280)\n\nHeaders:\n\n- Use either `Authorization: \u003CAPI_KEY>` or a short-lived `token` query parameter issued by your backend.\n\nMessages:\n\n- Client sends:\n  - Binary audio chunks (50–1000ms each).\n  - Optional JSON messages: `{\"type\": \"UpdateConfig\", ...}`, `{\"type\": \"Terminate\"}`, `{\"type\": \"ForceEndpoint\"}`.\n- Server sends:\n  - `Begin` event with `id`, `expires_at`.\n  - `Turn` events with:\n    - `transcript` (immutable partials\u002Ffinals),\n    - `utterance` (complete semantic chunk),\n    - `end_of_turn` (bool),\n    - `turn_is_formatted` (bool),\n    - `words` array with timestamps\u002Fconfidences.\n  - `Termination` event with summary stats.\n\n### 4. Regions and data residency\n\n- Async:\n  - US: `https:\u002F\u002Fapi.assemblyai.com`\n  - EU: `https:\u002F\u002Fapi.eu.assemblyai.com`\n- Streaming:\n  - US: `wss:\u002F\u002Fstreaming.assemblyai.com\u002Fv3\u002Fws`\n  - EU: `wss:\u002F\u002Fstreaming.eu.assemblyai.com\u002Fv3\u002Fws`\n\nAlways keep base URLs consistent per project; don’t mix US\u002FEU endpoints for the same data.\n\n---\n\n## Security & API Keys\n\n- Always require an AssemblyAI API key and keep it **out of source** in Claude Code output:\n  - Use environment variables: `ASSEMBLYAI_API_KEY`.\n  - Or placeholders (`\"\u003CYOUR_API_KEY>\"`) in snippets.\n- For browser\u002Fclient code:\n  - Do **not** embed the API key.\n  - Instruct the user to generate **temporary streaming tokens** on their backend and pass only the token into the WebSocket connection.\n- Never print real keys in logs or comments.\n\n---\n\n## High-Level Workflow Patterns\n\n### Decision tree\n\n1. **Is the audio live?**\n   - Yes → Use **Streaming STT**.\n   - No → Use **Async STT**.\n\n2. **Is latency critical (\u003C1s) for responses?**\n   - Yes → Streaming with `format_turns=false` and careful turn detection.\n   - No → Async, then Summarization\u002FChapters\u002Fetc.\n\n3. **Do transcripts leave the backend?**\n   - Yes → Consider `redact_pii` (and optionally `redact_pii_audio`) before sharing.\n   - No → Use raw transcripts as needed.\n\n4. **Need LLM-based processing (Q&A, structured summaries)?**\n   - Yes → Pipe transcripts into **LLM Gateway** via `chat\u002Fcompletions`.\n\n---\n\n## How Claude Should Work with This Skill\n\n### General principles\n\n- Prefer **official AssemblyAI SDKs** (Python\u002FJS) when available; fall back to `requests`\u002F`websocket-client` only if SDK cannot be installed.\n- Always:\n  - Validate HTTP responses and WebSocket status.\n  - Surface useful error messages (`status`, `error` fields in transcript JSON).\n  - Respect documented min\u002Fmax chunk sizes (50–1000ms of audio per binary message).\n- For voice-agent code, optimize for:\n  - Immutable partials (`transcript`) and `utterance` field.\n  - Minimal latency, avoid extra formatting passes.\n\n---\n\n## Recipe 1 – Minimal Streaming from Microphone (Python SDK)\n\n> Goal: Stream mic audio to AssemblyAI and print transcripts in real time.\n\nUse this when the environment has Python and `assemblyai` + `pyaudio` installed, and the user wants a quick streaming demo.\n\n```python\nimport assemblyai as aai\nfrom assemblyai.streaming import v3 as aai_stream\nimport pyaudio\n\nAPI_KEY = \"\u003CYOUR_API_KEY>\"\n\naai.settings.api_key = API_KEY\n\nSAMPLE_RATE = 16000\nCHUNK_MS = 50\nFRAMES_PER_BUFFER = int(SAMPLE_RATE * (CHUNK_MS \u002F 1000.0))\n\ndef main():\n    client = aai_stream.StreamingClient(\n        aai_stream.StreamingClientOptions(\n            api_key=API_KEY,\n            api_host=\"streaming.assemblyai.com\",  # or \"streaming.eu.assemblyai.com\"\n        )\n    )\n\n    def on_begin(_client, event: aai_stream.BeginEvent):\n        print(f\"Session started: {event.id}, expires at {event.expires_at}\")\n\n    def on_turn(_client, event: aai_stream.TurnEvent):\n        # Use immutable transcript text\n        text = (event.transcript or \"\").strip()\n        if not text:\n            return\n        # Use formatted finals only for display; keep unformatted for LLMs\n        if event.turn_is_formatted:\n            print(f\"[FINAL] {text}\")\n        else:\n            print(f\"[PARTIAL] {text}\", end=\"\\r\")\n\n    def on_terminated(_client, event: aai_stream.TerminationEvent):\n        print(f\"\\nTerminated. Audio duration={event.audio_duration_seconds}s\")\n\n    def on_error(_client, error: aai_stream.StreamingError):\n        print(f\"\\nStreaming error: {error}\")\n\n    client.on(aai_stream.StreamingEvents.Begin, on_begin)\n    client.on(aai_stream.StreamingEvents.Turn, on_turn)\n    client.on(aai_stream.StreamingEvents.Termination, on_terminated)\n    client.on(aai_stream.StreamingEvents.Error, on_error)\n\n    client.connect(\n        aai_stream.StreamingParameters(\n            sample_rate=SAMPLE_RATE,\n            format_turns=False,  # better latency for voice agents\n        )\n    )\n\n    pa = pyaudio.PyAudio()\n    stream = pa.open(\n        format=pyaudio.paInt16,\n        channels=1,\n        rate=SAMPLE_RATE,\n        input=True,\n        frames_per_buffer=FRAMES_PER_BUFFER,\n    )\n\n    try:\n        print(\"Speak into your microphone (Ctrl+C to stop)...\")\n        def audio_gen():\n            while True:\n                yield stream.read(FRAMES_PER_BUFFER, exception_on_overflow=False)\n        client.stream(audio_gen())\n    except KeyboardInterrupt:\n        pass\n    finally:\n        client.disconnect(terminate=True)\n        stream.stop_stream()\n        stream.close()\n        pa.terminate()\n\nif __name__ == \"__main__\":\n    main()\n","# AssemblyAI Streaming & Live Transcription Skill\n\n## Overview\n\nUse this skill to build and maintain code that talks to AssemblyAI’s:\n\n- **Streaming Speech-to-Text (STT)** via WebSockets (`wss:\u002F\u002Fstreaming.assemblyai.com\u002Fv3\u002Fws`)\n- **Async \u002F pre-recorded STT** via REST (`https:\u002F\u002Fapi.assemblyai.com\u002Fv2\u002Ftranscript`)\n- **LLM Gateway** for applying Claude\u002FGPT\u002FGemini-style models to transcripts (`https:\u002F\u002Fllm-gateway.assemblyai.com`)\n\nThe emphasis is on **streaming\u002Flive transcription**, **meeting notetakers**, and **voice agents**, while still covering async workflows and post-processing.\n\nThis skill assumes a Claude Code environment with access to Python (preferred) and Bash.\n\n---\n\n## When to Use\n\nUse this skill when:\n\n- Implementing **real-time transcription** from a microphone, telephony stream, or audio file.\n- Building a **live meeting notetaker** (Zoom\u002FTeams\u002FMeet), especially with summaries, action items, and highlights.\n- Implementing a **voice agent** where latency and natural turn-taking matter.\n- Migrating from other STT providers (OpenAI\u002FDeepgram\u002FGoogle\u002FAWS\u002Fetc.) to AssemblyAI.\n- Applying **LLMs to audio** via LLM Gateway for summaries, Q&A, topic tagging, or custom prompts.\n\nDo **not** use this skill when:\n\n- The task is generic HTTP client usage with no AssemblyAI-specific logic.\n- The request clearly targets a different STT vendor.\n- The environment cannot safely store or use an API key.\n\n---\n\n## AssemblyAI Mental Model\n\n### 1. Products to care about\n\n1. **Pre-recorded Speech-to-Text (Async)**\n   - REST API: `POST \u002Fv2\u002Ftranscript` → `GET \u002Fv2\u002Ftranscript\u002F{id}`\n   - Designed for files from URLs, uploads, S3, etc.\n   - Supports extra models: summarization, topic detection, sentiment, PII redaction, chapters, etc.\n\n2. **Streaming Speech-to-Text**\n   - WebSocket: `wss:\u002F\u002Fstreaming.assemblyai.com\u002Fv3\u002Fws`\n   - Low-latency, immutable transcripts (~300ms).\n   - Turn detection built in; fits voice agents and live captioning.\n\n3. **LLM Gateway**\n   - REST API: `POST \u002Fv1\u002Fchat\u002Fcompletions` at `https:\u002F\u002Fllm-gateway.assemblyai.com`\n   - Unified access to multiple LLMs (Claude, GPT, Gemini, etc.).\n   - Designed for “LLM over transcripts” workflows.\n\n### 2. Key model knobs (Async)\n\n- `speech_models`: `[\"slam-1\", \"universal\"]` etc.\n  - **Slam-1**: best English accuracy + keyterms_prompt, good for medical\u002Ftechnical conversations.\n  - **Universal**: multilingual coverage; good default if language is unknown.\n- `language_code` vs `language_detection`:\n  - Use `language_code` when the language is known.\n  - Use `language_detection: true` when unknown; optionally set `language_confidence_threshold`.\n- `keyterms_prompt`:\n  - Domain words\u002Fphrases to boost (med terms, product names, etc.).\n- Extra intelligence: `summarization`, `iab_categories`, `content_safety`, `entity_detection`, `auto_chapters`, `sentiment_analysis`, `speaker_labels`, `auto_highlights`, `redact_pii`, etc.\n\n### 3. Key model knobs (Streaming)\n\nConnection URL:\n\n- US: `wss:\u002F\u002Fstreaming.assemblyai.com\u002Fv3\u002Fws`\n- EU: `wss:\u002F\u002Fstreaming.eu.assemblyai.com\u002Fv3\u002Fws`\n\nImportant query parameters:\n\n- `sample_rate` (required): e.g. `16000`\n- `format_turns` (bool): return formatted final transcripts; **avoid** for low-latency voice agents.\n- `speech_model`: `universal-streaming-english` (default) or `universal-streaming-multi`.\n- `keyterms_p\n\nrompt`: JSON-encoded list of terms, e.g. `[\"AssemblyAI\", \"Slam-1\", \"Keanu Reeves\"]`.\n- Turn detection:\n  - `end_of_turn_confidence_threshold` (0.0–1.0, default ~0.4)\n  - `min_end_of_turn_silence_when_confident` (ms, default ~400)\n  - `max_turn_silence` (ms, default ~1280)\n\nHeaders:\n\n- Use either `Authorization: \u003CAPI_KEY>` or a short-lived `token` query parameter issued by your backend.\n\nMessages:\n\n- Client sends:\n  - Binary audio chunks (50–1000ms each).\n  - Optional JSON messages: `{\"type\": \"UpdateConfig\", ...}`, `{\"type\": \"Terminate\"}`, `{\"type\": \"ForceEndpoint\"}`.\n- Server sends:\n  - `Begin` event with `id`, `expires_at`.\n  - `Turn` events with:\n    - `transcript` (immutable partials\u002Ffinals),\n    - `utterance` (complete semantic chunk),\n    - `end_of_turn` (bool),\n    - `turn_is_formatted` (bool),\n    - `words` array with timestamps\u002Fconfidences.\n  - `Termination` event with summary stats.\n\n### 4. Regions and data residency\n\n- Async:\n  - US: `https:\u002F\u002Fapi.assemblyai.com`\n  - EU: `https:\u002F\u002Fapi.eu.assemblyai.com`\n- Streaming:\n  - US: `wss:\u002F\u002Fstreaming.assemblyai.com\u002Fv3\u002Fws`\n  - EU: `wss:\u002F\u002Fstreaming.eu.assemblyai.com\u002Fv3\u002Fws`\n\nAlways keep base URLs consistent per project; don’t mix US\u002FEU endpoints for the same data.\n\n---\n\n## Security & API Keys\n\n- Always require an AssemblyAI API key and keep it **out of source** in Claude Code output:\n  - Use environment variables: `ASSEMBLYAI_API_KEY`.\n  - Or placeholders (`\"\u003CYOUR_API_KEY>\"`) in snippets.\n- For browser\u002Fclient code:\n  - Do **not** embed the API key.\n  - Instruct the user to generate **temporary streaming tokens** on their backend and pass only the token into the WebSocket connection.\n- Never print real keys in logs or comments.\n\n---\n\n## High-Level Workflow Patterns\n\n### Decision tree\n\n1. **Is the audio live?**\n   - Yes → Use **Streaming STT**.\n   - No → Use **Async STT**.\n\n2. **Is latency critical (\u003C1s) for responses?**\n   - Yes → Streaming with `format_turns=false` and careful turn detection.\n   - No → Async, then Summarization\u002FChapters\u002Fetc.\n\n3. **Do transcripts leave the backend?**\n   - Yes → Consider `redact_pii` (and optionally `redact_pii_audio`) before sharing.\n   - No → Use raw transcripts as needed.\n\n4. **Need LLM-based processing (Q&A, structured summaries)?**\n   - Yes → Pipe transcripts into **LLM Gateway** via `chat\u002Fcompletions`.\n\n---\n\n## How Claude Should Work with This Skill\n\n### General principles\n\n- Prefer **official AssemblyAI SDKs** (Python\u002FJS) when available; fall back to `requests`\u002F`websocket-client` only if SDK cannot be installed.\n- Always:\n  - Validate HTTP responses and WebSocket status.\n  - Surface useful error messages (`status`, `error` fields in transcript JSON).\n  - Respect documented min\u002Fmax chunk sizes (50–1000ms of audio per binary message).\n- For voice-agent code, optimize for:\n  - Immutable partials (`transcript`) and `utterance` field.\n  - Minimal latency, avoid extra formatting passes.\n\n---\n\n## Recipe 1 – Minimal Streaming from Microphone (Python SDK)\n\n> Goal: Stream mic audio to AssemblyAI and print transcripts in real time.\n\nUse this when the environment has Python and `assemblyai` + `pyaudio` installed, and the user wants a quick streaming demo.\n\n```python\nimport assemblyai as aai\nfrom assemblyai.streaming import v3 as aai_stream\nimport pyaudio\n\nAPI_KEY = \"\u003CYOUR_API_KEY>\"\n\naai.settings.api_key = API_KEY\n\nSAMPLE_RATE = 16000\nCHUNK_MS = 50\nFRAMES_PER_BUFFER = int(SAMPLE_RATE * (CHUNK_MS \u002F 1000.0))\n\ndef main():\n    client = aai_stream.StreamingClient(\n        aai_stream.StreamingClientOptions(\n            api_key=API_KEY,\n            api_host=\"streaming.assemblyai.com\",  # or \"streaming.eu.assemblyai.com\"\n        )\n    )\n\n    def on_begin(_client, event: aai_stream.BeginEvent):\n        print(f\"Session started: {event.id}, expires at {event.expires_at}\")\n\n    def on_turn(_client, event: aai_stream.TurnEvent):\n        # Use immutable transcript text\n        text = (event.transcript or \"\").strip()\n        if not text:\n            return\n        # Use formatted finals only for display; keep unformatted for LLMs\n        if event.turn_is_formatted:\n            print(f\"[FINAL] {text}\")\n        else:\n            print(f\"[PARTIAL] {text}\", end=\"\\r\")\n\n    def on_terminated(_client, event: aai_stream.TerminationEvent):\n        print(f\"\\nTerminated. Audio duration={event.audio_duration_seconds}s\")\n\n    def on_error(_client, error: aai_stream.StreamingError):\n        print(f\"\\nStreaming error: {error}\")\n\n    client.on(aai_stream.StreamingEvents.Begin, on_begin)\n    client.on(aai_stream.StreamingEvents.Turn, on_turn)\n    client.on(aai_stream.StreamingEvents.Termination, on_terminated)\n    client.on(aai_stream.StreamingEvents.Error, on_error)\n\n    client.connect(\n        aai_stream.StreamingParameters(\n            sample_rate=SAMPLE_RATE,\n            format_turns=False,  # better latency for voice agents\n        )\n    )\n\n    pa = pyaudio.PyAudio()\n    stream = pa.open(\n        format=pyaudio.paInt16,\n        channels=1,\n        rate=SAMPLE_RATE,\n        input=True,\n        frames_per_buffer=FRAMES_PER_BUFFER,\n    )\n\n    try:\n        print(\"Speak into your microphone (Ctrl+C to stop)...\")\n        def audio_gen():\n            while True:\n                yield stream.read(FRAMES_PER_BUFFER, exception_on_overflow=False)\n        client.stream(audio_gen())\n    except KeyboardInterrupt:\n        pass\n    finally:\n        client.disconnect(terminate=True)\n        stream.stop_stream()\n        stream.close()\n        pa.terminate()\n\nif __name__ == \"__main__\":\n    main()\n",{"name":5,"license":15,"metadata":54,"description":7,"allowed-tools":58},{"focus":55,"skill-version":56,"upstream-docs":57},"streaming-stt, meeting-notetaker, voice-agent, llm-gateway","1.0.0","https:\u002F\u002Fwww.assemblyai.com\u002Fdocs",[59,60,61,62,63,43,64],"Read","Write","Edit","Grep","Glob","Python","main","fa638b97a18ff6d9f26cf0eaf3dfa123531bcc87","organized-skills\u002Fassembly-ai-streaming",1,"https:\u002F\u002Fgithub.com\u002Flevalencia\u002Fagent-god-mode","https:\u002F\u002Fgithub.com\u002Flevalencia","https:\u002F\u002Fmedium.com\u002F@luisevalencia","Microsoft MVP - Artificial Intelligence\r\nMachine Learning Enthusiast\r\n","2026-05-02 06:55:36",{"source":8,"skillName":5,"commands":75,"prompt":85,"zipAvailable":86},[76,79,82],{"tool":77,"command":78},"npx","npx skills add levalencia\u002Fagent-god-mode --skill assemblyai-streaming",{"tool":80,"command":81},"bunx","bunx skills add levalencia\u002Fagent-god-mode --skill assemblyai-streaming",{"tool":83,"command":84},"pnpm","pnpm dlx skills add levalencia\u002Fagent-god-mode --skill assemblyai-streaming","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Flevalencia\u002Fagent-god-mode\u002Fblob\u002Fmain\u002Forganized-skills\u002Fassembly-ai-streaming\u002FSKILL.md",true,[88,94,100,104,108,112,116],{"path":89,"resourceType":90,"resourceTypeDesc":91,"sizeBytes":92,"sourceHtmlUrl":93},"LICENSE.txt","LICENSE","许可证",1069,"https:\u002F\u002Fgithub.com\u002Fmonoes\u002Fmonomind\u002Fblob\u002Fmain\u002Fpackages\u002F@monomind\u002Fcli\u002Forg-skills\u002Fsales-prospecting\u002FLICENSE.txt",{"path":95,"resourceType":96,"resourceTypeDesc":97,"sizeBytes":98,"sourceHtmlUrl":99},"references\u002Fb2b-prospecting.md","REFERENCE","参考文档",4960,"https:\u002F\u002Fgithub.com\u002Fmonoes\u002Fmonomind\u002Fblob\u002Fmain\u002Fpackages\u002F@monomind\u002Fcli\u002Forg-skills\u002Fsales-prospecting\u002Freferences\u002Fb2b-prospecting.md",{"path":101,"resourceType":96,"resourceTypeDesc":97,"sizeBytes":102,"sourceHtmlUrl":103},"references\u002Fcompliance.md",6232,"https:\u002F\u002Fgithub.com\u002Fmonoes\u002Fmonomind\u002Fblob\u002Fmain\u002Fpackages\u002F@monomind\u002Fcli\u002Forg-skills\u002Fsales-prospecting\u002Freferences\u002Fcompliance.md",{"path":105,"resourceType":96,"resourceTypeDesc":97,"sizeBytes":106,"sourceHtmlUrl":107},"references\u002Fdata-sources.md",11400,"https:\u002F\u002Fgithub.com\u002Fmonoes\u002Fmonomind\u002Fblob\u002Fmain\u002Fpackages\u002F@monomind\u002Fcli\u002Forg-skills\u002Fsales-prospecting\u002Freferences\u002Fdata-sources.md",{"path":109,"resourceType":96,"resourceTypeDesc":97,"sizeBytes":110,"sourceHtmlUrl":111},"references\u002Fdemand-signals.md",9985,"https:\u002F\u002Fgithub.com\u002Fmonoes\u002Fmonomind\u002Fblob\u002Fmain\u002Fpackages\u002F@monomind\u002Fcli\u002Forg-skills\u002Fsales-prospecting\u002Freferences\u002Fdemand-signals.md",{"path":113,"resourceType":96,"resourceTypeDesc":97,"sizeBytes":114,"sourceHtmlUrl":115},"references\u002Flocal-prospecting.md",8150,"https:\u002F\u002Fgithub.com\u002Fmonoes\u002Fmonomind\u002Fblob\u002Fmain\u002Fpackages\u002F@monomind\u002Fcli\u002Forg-skills\u002Fsales-prospecting\u002Freferences\u002Flocal-prospecting.md",{"path":117,"resourceType":96,"resourceTypeDesc":97,"sizeBytes":118,"sourceHtmlUrl":119},"references\u002Fsaas-prospecting.md",6051,"https:\u002F\u002Fgithub.com\u002Fmonoes\u002Fmonomind\u002Fblob\u002Fmain\u002Fpackages\u002F@monomind\u002Fcli\u002Forg-skills\u002Fsales-prospecting\u002Freferences\u002Fsaas-prospecting.md",[121,142,174,200,225,253],{"skillNo":122,"name":123,"nameZh":6,"description":124,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":125,"categories":126,"keywords":127,"publishedAt":49},"01M6EM8ES5M9J4S8N78NJST0YC","zarr-python","Chunked N-D arrays for cloud storage. Compressed arrays, parallel I\u002FO, S3\u002FGCS integration, NumPy\u002FDask\u002FXarray compatible, for large-scale scientific computing pipelines.","https:\u002F\u002Fgithub.com\u002Flevalencia\u002Fagent-god-mode\u002Fblob\u002Fmain\u002Forganized-skills\u002Fzarr-python\u002FSKILL.md",[],[40,128,129,130,131,132,133,134,135,136,137,138,139,140,141],"zarr python","storage","integration","s3","gcs","cloud","dask","parallel","pipelines","arrays","numpy","scientific computing","scale","scientific",{"skillNo":143,"name":144,"nameZh":6,"description":145,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":146,"categories":147,"keywords":155,"publishedAt":49},"01M6EM8AR6ZFKZ9YVW8TTG850Z","zabbix","Zabbix monitoring system automation via API and Python. Use when: (1) Managing hosts, templates, items, triggers, or host groups, (2) Automating monitoring configuration, (3) Sending data via Zabbix trapper\u002Fsender, (4) Querying historical data or events, (5) Bulk operations on Zabbix objects, (6) Maintenance window management, (7) User\u002Fpermission management","https:\u002F\u002Fgithub.com\u002Flevalencia\u002Fagent-god-mode\u002Fblob\u002Fmain\u002Forganized-skills\u002Fzabbix-api-skill\u002FSKILL.md",[148,151,152],{"categoryNo":149,"categoryName":150,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000010","后端与接口",{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":153,"categoryName":154,"taxonomyType":26,"taxonomyTypeDesc":27},"01JBSEEDCAT000000000000030","开发工程师",[144,40,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173],"automation","window management","monitoring","data","templates","api","system","operations","use","maintenance","objects","permission","bulk operations","triggers","host","events","via","bulk",{"skillNo":175,"name":176,"nameZh":6,"description":177,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":178,"categories":179,"keywords":180,"publishedAt":49},"01M6EM6Q9JG846ECQY9A210A26","x-twitter-growth","X\u002FTwitter growth engine for building audience, crafting viral content, and analyzing engagement. Use when the user wants to grow on X\u002FTwitter, write tweets or threads, analyze their X profile, research competitors on X, plan a posting strategy, or optimize engagement. Complements social-content (generic multi-platform) with X-specific depth: algorithm mechanics, thread engineering, reply strategy, profile optimization, and competitive intelligence via web search.","https:\u002F\u002Fgithub.com\u002Flevalencia\u002Fagent-god-mode\u002Fblob\u002Fmain\u002Forganized-skills\u002Fx-twitter-growth\u002FSKILL.md",[],[181,182,183,184,185,186,187,188,189,190,191,37,192,193,194,195,196,197,198,199],"twitter","x twitter growth","growth","competitive intelligence","search","analyzing","plan","marketing","web search","competitors","analyze","optimize","social","intelligence","viral","growth engine","mechanics","grow","competitive",{"skillNo":201,"name":202,"nameZh":6,"description":203,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":204,"categories":205,"keywords":206,"publishedAt":49},"01M6EM3FQNPYNKPC4CV9TCVNGD","what-if-oracle","Run structured What-If scenario analysis with multi-branch possibility exploration. Use this skill when the user asks speculative questions like \"what if...\", \"what would happen if...\", \"what are the possibilities\", \"explore scenarios\", \"scenario analysis\", \"possibility space\", \"what could go wrong\", \"best case \u002F worst case\", \"risk analysis\", \"contingency planning\", \"strategic options\", or any question about uncertain futures. Also trigger when the user faces a fork-in-the-road decision, wants to stress-test an idea, or needs to think through consequences before committing.","https:\u002F\u002Fgithub.com\u002Flevalencia\u002Fagent-god-mode\u002Fblob\u002Fmain\u002Forganized-skills\u002Fwhat-if-oracle\u002FSKILL.md",[],[207,208,209,210,211,212,213,45,214,215,216,217,164,218,219,220,221,222,223,224],"oracle","what if oracle","test","planning","ux","futures","risk","explore","decision","branch","question","exploration","options","stress-test","strategies","trigger","case","multi",{"skillNo":226,"name":227,"nameZh":6,"description":228,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":229,"categories":230,"keywords":234,"publishedAt":49},"01M6EM08C06RGDYCJ86HAPK0FF","vercel-react-best-practices","React and Next.js performance optimization guidelines from Vercel Engineering. This skill should be used when writing, reviewing, or refactoring React\u002FNext.js code to ensure optimal performance patterns. Triggers on tasks involving React components, Next.js pages, data fetching, bundle optimization, or performance improvements.","https:\u002F\u002Fgithub.com\u002Flevalencia\u002Fagent-god-mode\u002Fblob\u002Fmain\u002Forganized-skills\u002Fvercel-react-best-practices\u002FSKILL.md",[231],{"categoryNo":232,"categoryName":233,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000009","前端开发",[235,236,237,238,239,240,241,242,243,244,245,246,247,159,248,249,250,251,252,169],"react","vercel react best practices","best practices","react best practices","vercel","writing","performance","components","react components","guidelines","tasks","code","Optimization","Next.js","refactoring","performance optimization","next","pages",{"skillNo":254,"name":255,"nameZh":6,"description":256,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":257,"categories":258,"keywords":269,"publishedAt":49},"01M6EM00S13326QM08DTM94J6Z","venue-templates","Access comprehensive LaTeX templates, formatting requirements, and submission guidelines for major scientific publication venues (Nature, Science, PLOS, IEEE, ACM), academic conferences (NeurIPS, ICML, CVPR, CHI), research posters, and grant proposals (NSF, NIH, DOE, DARPA). This skill should be used when preparing manuscripts for journal submission, conference papers, research posters, or grant proposals and need venue-specific formatting requirements and templates.","https:\u002F\u002Fgithub.com\u002Flevalencia\u002Fagent-god-mode\u002Fblob\u002Fmain\u002Forganized-skills\u002Fvenue-templates\u002FSKILL.md",[259,262,265,266],{"categoryNo":260,"categoryName":261,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000018","写作与文案",{"categoryNo":263,"categoryName":264,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000021","资料检索与研究",{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":267,"categoryName":268,"taxonomyType":26,"taxonomyTypeDesc":27},"01JBSEEDCAT000000000000036","内容创作者",[160,270,271,244,272,273,274,275,43,276,38,37,36,277,278,279,280,281,141,282],"venue templates","research","NeurIPS","ICML","LaTeX","science","access","proposals","Academic","requirements","formatting","journal","chi"]