[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M77NWEHN1YDVC6WW673HTT5N":3,"skill-same:01M77NWEHN1YDVC6WW673HTT5N":60},{"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":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":37,"commitSha":38,"skillDirectory":39,"forksCount":40,"repositoryHtmlUrl":41,"repositoryHomepageUrl":42,"ownerHtmlUrl":43,"ownerBlogUrl":6,"ownerDescription":44,"updatedAt":45,"install":46,"resources":59},"01M77NWEHN1YDVC6WW673HTT5N","natural-language-video-search",null,"Semantic search over video files using Gemini embeddings. Index dashcam, security camera, or any mp4 footage, then search with natural language queries to find and auto-trim matching clips.","Lord1Egypt\u002Fawesome-skill-forge","Lord1Egypt","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F99746359?v=4","01M7592YDNT0XK9WJ0T2MQVPDN","Mohamed Mounir Ali AL Keshita","01M75GRFNHC8VZNWEB8DKDE433",3,"MIT","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fcommunity\u002Fclawhub\u002Fn\u002Fnatural-language-video-search\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: natural-language-video-search\ndescription: >\n  Semantic search over video files using Gemini embeddings.\n  Index dashcam, security camera, or any mp4 footage, then search\n  with natural language queries to find and auto-trim matching clips.\nversion: 0.2.3\nmetadata:\n  clawdbot:\n    requires:\n      env:\n        - GEMINI_API_KEY\n      bins:\n        - python3\n        - uv\n        - ffmpeg\n    primaryEnv: GEMINI_API_KEY\n    homepage: https:\u002F\u002Fgithub.com\u002Fssrajadh\u002Fsentrysearch\n    emoji: \"🎥\"\n---\n\n# Natural Language Video Search\n\nSearch video files using natural language queries powered by Gemini Embedding 2's native video-to-vector embedding.\n\n## What This Skill Does\n\nThis skill lets you index video files (dashcam footage, security camera recordings, any mp4) into a local vector database, then search them by describing what you're looking for in plain English. The top match is automatically trimmed and saved as a clip.\n\nFor Tesla dashcam footage, an optional telemetry overlay can burn speed, GPS, location, and turn signal data onto trimmed clips.\n\n## Setup\n\nRequires [uv](https:\u002F\u002Fdocs.astral.sh\u002Fuv\u002Fgetting-started\u002Finstallation\u002F) and Python 3.11+.\n\n1. Clone and install:\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fssrajadh\u002Fsentrysearch.git\ncd sentrysearch\nuv sync\n```\n\nFor Tesla overlay support (reverse geocoding):\n\n```bash\nuv sync --extra tesla\n```\n\n2. Set your Gemini API key:\n\n```bash\nsentrysearch init\n```\n\nThis prompts for your key, writes it to `.env`, and validates it with a test embedding. You can also set `GEMINI_API_KEY` directly as an environment variable.\n\n## Commands\n\n### Index video files\n\n```bash\nsentrysearch index \u003Cdirectory_or_file>\n```\n\nOptions: `--chunk-duration` (default 30s), `--overlap` (default 5s), `--no-preprocess`, `--target-resolution`, `--target-fps`, `--skip-still` \u002F `--no-skip-still`, `--verbose`\n\n### Search indexed footage\n\n```bash\nsentrysearch search \"\u003Cnatural language query>\"\n```\n\nOptions: `-n` \u002F `--results` (default 5), `-o` \u002F `--output-dir`, `--trim` \u002F `--no-trim`, `--threshold` (default 0.41), `--overlay` \u002F `--no-overlay` (Tesla telemetry), `--verbose`\n\n### Apply Tesla telemetry overlay\n\n```bash\nsentrysearch overlay \u003Cvideo_file>\nsentrysearch overlay \u003Cvideo_file> -o output.mp4\n```\n\nBurns a HUD overlay onto a Tesla dashcam video showing speed, GPS coordinates, location name, and turn signal status. Reads telemetry from SEI NAL units embedded in Tesla firmware 2025.44.25+. Also available as `--overlay` flag on the search command to automatically overlay the trimmed clip.\n\n### Check index stats\n\n```bash\nsentrysearch stats\n```\n\n## How It Works\n\nVideo files are split into overlapping chunks. Still-frame detection can skip chunks with no meaningful visual change, eliminating unnecessary API calls — this is the primary cost saver for idle footage like sentry mode or security cameras. Chunks are also preprocessed (reduced frame rate and resolution) to shrink upload size and speed up transfers, though the Gemini API bills based on video duration at a fixed token rate, not file size, so preprocessing does not reduce per-chunk token cost. Each chunk is embedded as raw video using Gemini Embedding 2 (no transcription or captioning). Vectors are stored in a local ChromaDB database. Text queries are embedded into the same vector space and matched via cosine similarity. The top match is auto-trimmed from the original file via ffmpeg.\n\n## When To Use This Skill\n\n- User asks to search through video files or footage\n- User wants to find a specific moment in a video by describing it\n- User asks to index or organize video footage for search\n- User mentions dashcam, security camera, or surveillance clips\n- User wants to find and extract a clip from a longer video\n- User has Tesla dashcam footage and wants speed\u002FGPS\u002Flocation overlay on clips\n- User wants to apply telemetry overlay to a Tesla video\n\n## Example Interactions\n\nUser: \"Search my dashcam footage for a white truck cutting me off\"\nAction: Run `sentrysearch search \"white truck cutting me off\"`\n\nUser: \"Index all the video files in my Downloads folder\"\nAction: Run `sentrysearch index ~\u002FDownloads`\n\nUser: \"Search for a red light and include the Tesla overlay on the clip\"\nAction: Run `sentrysearch search \"running a red light\" --overlay`\n\nUser: \"Add the speed and GPS overlay to this Tesla video\"\nAction: Run `sentrysearch overlay \u002Fpath\u002Fto\u002Ftesla_video.mp4`\n\nUser: \"How much footage do I have indexed?\"\nAction: Run `sentrysearch stats`\n\n## Rules\n\n- Always run `sentrysearch init` or confirm GEMINI_API_KEY is set before indexing or searching.\n- If ffmpeg is not found on PATH, the bundled `imageio-ffmpeg` fallback is used automatically.\n- Indexing costs ~$2.84\u002Fhour of active footage with default settings. Cost is driven by the number of chunks sent to the API — footage with long idle periods (sentry mode, security cameras) will be significantly cheaper since still-frame skipping eliminates those chunks entirely. Warn the user before indexing large directories.\n- Search results include similarity scores. Scores below the threshold (default 0.41) trigger a low-confidence prompt before trimming.\n- The Tesla overlay requires firmware 2025.44.25+ for SEI metadata. Videos without Tesla metadata will skip the overlay gracefully.\n- Requires Python 3.11+.\n","# Natural Language Video Search\n\nSearch video files using natural language queries powered by Gemini Embedding 2's native video-to-vector embedding.\n\n## What This Skill Does\n\nThis skill lets you index video files (dashcam footage, security camera recordings, any mp4) into a local vector database, then search them by describing what you're looking for in plain English. The top match is automatically trimmed and saved as a clip.\n\nFor Tesla dashcam footage, an optional telemetry overlay can burn speed, GPS, location, and turn signal data onto trimmed clips.\n\n## Setup\n\nRequires [uv](https:\u002F\u002Fdocs.astral.sh\u002Fuv\u002Fgetting-started\u002Finstallation\u002F) and Python 3.11+.\n\n1. Clone and install:\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fssrajadh\u002Fsentrysearch.git\ncd sentrysearch\nuv sync\n```\n\nFor Tesla overlay support (reverse geocoding):\n\n```bash\nuv sync --extra tesla\n```\n\n2. Set your Gemini API key:\n\n```bash\nsentrysearch init\n```\n\nThis prompts for your key, writes it to `.env`, and validates it with a test embedding. You can also set `GEMINI_API_KEY` directly as an environment variable.\n\n## Commands\n\n### Index video files\n\n```bash\nsentrysearch index \u003Cdirectory_or_file>\n```\n\nOptions: `--chunk-duration` (default 30s), `--overlap` (default 5s), `--no-preprocess`, `--target-resolution`, `--target-fps`, `--skip-still` \u002F `--no-skip-still`, `--verbose`\n\n### Search indexed footage\n\n```bash\nsentrysearch search \"\u003Cnatural language query>\"\n```\n\nOptions: `-n` \u002F `--results` (default 5), `-o` \u002F `--output-dir`, `--trim` \u002F `--no-trim`, `--threshold` (default 0.41), `--overlay` \u002F `--no-overlay` (Tesla telemetry), `--verbose`\n\n### Apply Tesla telemetry overlay\n\n```bash\nsentrysearch overlay \u003Cvideo_file>\nsentrysearch overlay \u003Cvideo_file> -o output.mp4\n```\n\nBurns a HUD overlay onto a Tesla dashcam video showing speed, GPS coordinates, location name, and turn signal status. Reads telemetry from SEI NAL units embedded in Tesla firmware 2025.44.25+. Also available as `--overlay` flag on the search command to automatically overlay the trimmed clip.\n\n### Check index stats\n\n```bash\nsentrysearch stats\n```\n\n## How It Works\n\nVideo files are split into overlapping chunks. Still-frame detection can skip chunks with no meaningful visual change, eliminating unnecessary API calls — this is the primary cost saver for idle footage like sentry mode or security cameras. Chunks are also preprocessed (reduced frame rate and resolution) to shrink upload size and speed up transfers, though the Gemini API bills based on video duration at a fixed token rate, not file size, so preprocessing does not reduce per-chunk token cost. Each chunk is embedded as raw video using Gemini Embedding 2 (no transcription or captioning). Vectors are stored in a local ChromaDB database. Text queries are embedded into the same vector space and matched via cosine similarity. The top match is auto-trimmed from the original file via ffmpeg.\n\n## When To Use This Skill\n\n- User asks to search through video files or footage\n- User wants to find a specific moment in a video by describing it\n- User asks to index or organize video footage for search\n- User mentions dashcam, security camera, or surveillance clips\n- User wants to find and extract a clip from a longer video\n- User has Tesla dashcam footage and wants speed\u002FGPS\u002Flocation overlay on clips\n- User wants to apply telemetry overlay to a Tesla video\n\n## Example Interactions\n\nUser: \"Search my dashcam footage for a white truck cutting me off\"\nAction: Run `sentrysearch search \"white truck cutting me off\"`\n\nUser: \"Index all the video files in my Downloads folder\"\nAction: Run `sentrysearch index ~\u002FDownloads`\n\nUser: \"Search for a red light and include the Tesla overlay on the clip\"\nAction: Run `sentrysearch search \"running a red light\" --overlay`\n\nUser: \"Add the speed and GPS overlay to this Tesla video\"\nAction: Run `sentrysearch overlay \u002Fpath\u002Fto\u002Ftesla_video.mp4`\n\nUser: \"How much footage do I have indexed?\"\nAction: Run `sentrysearch stats`\n\n## Rules\n\n- Always run `sentrysearch init` or confirm GEMINI_API_KEY is set before indexing or searching.\n- If ffmpeg is not found on PATH, the bundled `imageio-ffmpeg` fallback is used automatically.\n- Indexing costs ~$2.84\u002Fhour of active footage with default settings. Cost is driven by the number of chunks sent to the API — footage with long idle periods (sentry mode, security cameras) will be significantly cheaper since still-frame skipping eliminates those chunks entirely. Warn the user before indexing large directories.\n- Search results include similarity scores. Scores below the threshold (default 0.41) trigger a low-confidence prompt before trimming.\n- The Tesla overlay requires firmware 2025.44.25+ for SEI metadata. Videos without Tesla metadata will skip the overlay gracefully.\n- Requires Python 3.11+.\n",{"name":5,"version":24,"metadata":25,"description":36},"0.2.3",{"clawdbot":26},{"emoji":27,"homepage":28,"requires":29,"primaryEnv":31},"🎥","https:\u002F\u002Fgithub.com\u002Fssrajadh\u002Fsentrysearch",{"env":30,"bins":32},[31],"GEMINI_API_KEY",[33,34,35],"python3","uv","ffmpeg","Semantic search over video files using Gemini embeddings. Index dashcam, security camera, or any mp4 footage, then search with natural language queries to find and auto-trim matching clips.\n","master","4aa004da461d820f4ea5499e58eb8d196cf99016","community\u002Fclawhub\u002Fn\u002Fnatural-language-video-search",0,"https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge","https:\u002F\u002Fskill-explorer-rho.vercel.app","https:\u002F\u002Fgithub.com\u002FLord1Egypt","Unbreakable. Unstoppable. Unmatched⚡.\r\n Speaker of machine tongues 🤖  | World-builder of decentralized dreams 🌐 | Defender of the digital realm🛡️.","2026-06-10 08:21:01",{"source":8,"skillName":5,"commands":47,"prompt":57,"zipAvailable":58},[48,51,54],{"tool":49,"command":50},"npx","npx skills add Lord1Egypt\u002Fawesome-skill-forge --skill natural-language-video-search",{"tool":52,"command":53},"bunx","bunx skills add Lord1Egypt\u002Fawesome-skill-forge --skill natural-language-video-search",{"tool":55,"command":56},"pnpm","pnpm dlx skills add Lord1Egypt\u002Fawesome-skill-forge --skill natural-language-video-search","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fcommunity\u002Fclawhub\u002Fn\u002Fnatural-language-video-search\u002FSKILL.md",true,[],[61,68,75,82,89,96],{"skillNo":62,"name":63,"nameZh":6,"description":64,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":65,"categories":66,"keywords":67,"publishedAt":19},"01M785WKJB99G3STR1NF9NRTJW","youtube-content","YouTube transcripts to summaries, threads, blogs.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Fyoutube-content\u002FSKILL.md",[],[],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":19},"01M785WEFHB0G2S9YA6EPR3H0E","xurl","X\u002FTwitter via xurl CLI: post, search, DM, media, v2 API.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Fxurl\u002FSKILL.md",[],[],{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":19},"01M785W98EDN4EFJXJPMQEHCRC","touchdesigner-mcp","Control a running TouchDesigner instance via twozero MCP — create operators, set parameters, wire connections, execute Python, build real-time visuals. 36 native tools.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Ftouchdesigner-mcp\u002FSKILL.md",[],[],{"skillNo":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"publishedAt":19},"01M785W5S9FP3QBH2V191E025P","test-driven-development","TDD: enforce RED-GREEN-REFACTOR, tests before code.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Ftest-driven-development\u002FSKILL.md",[],[],{"skillNo":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":95,"publishedAt":19},"01M785W2AS08NSSH9YATCV6FJB","teams-meeting-pipeline","Operate the Teams meeting summary pipeline via Hermes CLI — summarize meetings, inspect pipeline status, replay jobs, manage Microsoft Graph subscriptions.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Fteams-meeting-pipeline\u002FSKILL.md",[],[],{"skillNo":97,"name":98,"nameZh":6,"description":99,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":100,"categories":101,"keywords":102,"publishedAt":19},"01M785VXKW4T34BTP1SCHW1R7F","systematic-debugging","4-phase root cause debugging: understand bugs before fixing.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Fsystematic-debugging\u002FSKILL.md",[],[]]