{"id":50325304,"url":"https://github.com/testy-cool/video-analyzer-ai","last_synced_at":"2026-05-29T05:30:59.250Z","repository":{"id":356807189,"uuid":"1231087820","full_name":"testy-cool/video-analyzer-ai","owner":"testy-cool","description":null,"archived":false,"fork":false,"pushed_at":"2026-05-09T20:10:21.000Z","size":65,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-05-09T22:07:09.368Z","etag":null,"topics":["ai","cli","llm","openrouter","tiktok","transcript","video-analysis","video-to-text","youtube-transcript","yt-dlp"],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/testy-cool.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-05-06T16:02:27.000Z","updated_at":"2026-05-09T20:10:25.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/testy-cool/video-analyzer-ai","commit_stats":null,"previous_names":["testy-cool/video-analyzer-ai"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/testy-cool/video-analyzer-ai","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/testy-cool%2Fvideo-analyzer-ai","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/testy-cool%2Fvideo-analyzer-ai/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/testy-cool%2Fvideo-analyzer-ai/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/testy-cool%2Fvideo-analyzer-ai/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/testy-cool","download_url":"https://codeload.github.com/testy-cool/video-analyzer-ai/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/testy-cool%2Fvideo-analyzer-ai/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":33639050,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-05-29T02:00:06.066Z","response_time":107,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai","cli","llm","openrouter","tiktok","transcript","video-analysis","video-to-text","youtube-transcript","yt-dlp"],"created_at":"2026-05-29T05:30:58.669Z","updated_at":"2026-05-29T05:30:59.245Z","avatar_url":"https://github.com/testy-cool.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# video-analyzer-ai\n\nExtract transcripts and analyze videos with AI. YouTube, Twitter/X, TikTok, Instagram, and [1000+ sites](https://github.com/yt-dlp/yt-dlp/blob/master/supportedsites.md) supported.\n\n```bash\nva \"https://www.youtube.com/watch?v=dQw4w9WgXcQ\"\nva \"https://x.com/user/status/123456\" -a\nva \"https://www.tiktok.com/@user/video/123456\" -a -v\n```\n\n## How it works\n\n```\nURL → YouTube captions (free) → transcript\n        ↓ (no captions?)\n      yt-dlp audio download → multimodal LLM transcription → transcript\n        ↓ (--analyze?)\n      LLM analysis via any OpenAI-compatible API → structured breakdown\n        ↓ (--verify?)\n      Second pass with prompt cache optimization → verified result\n```\n\n- **YouTube**: tries built-in captions first (free, instant), falls back to audio transcription\n- **Everything else**: downloads audio via yt-dlp, transcribes it with a multimodal LLM (Gemini by default)\n- **Analysis**: sends transcript to any OpenAI-compatible endpoint (OpenRouter, OpenAI, Gemini, local models)\n- **Verification**: double-pass analysis where the second call reuses the prompt cache (~66% cheaper)\n- **Caching**: transcripts, metadata, and analysis results cached locally with human-readable filenames\n\n## Install\n\nRequires Python 3.13+ and [yt-dlp](https://github.com/yt-dlp/yt-dlp).\n\n```bash\n# Install yt-dlp if you don't have it\npip install yt-dlp\n\n# Install video-analyzer-ai\ngit clone https://github.com/testy-cool/video-analyzer-ai.git\ncd video-analyzer-ai\ncp .env.example .env  # add your API keys\npip install -e .\n```\n\nOr with [uv](https://docs.astral.sh/uv/):\n\n```bash\nuv tool install --editable .\n```\n\nThe CLI command is `va`.\n\n## Configuration\n\nCopy `.env.example` to `.env` and fill in your keys:\n\n```bash\n# Analysis + transcription LLM — any OpenAI-compatible endpoint\nANALYSIS_BASE_URL=https://openrouter.ai/api/v1/chat/completions\nANALYSIS_API_KEY=sk-...\nANALYSIS_MODEL=google/gemini-3.1-flash-lite\n\n# Transcription model — must accept audio input (multimodal)\nTRANSCRIPTION_MODEL=google/gemini-3.1-flash-lite\n\n# Optional: Langfuse tracing\nLANGFUSE_SECRET_KEY=\nLANGFUSE_PUBLIC_KEY=\nLANGFUSE_BASE_URL=\n\n# Optional: Evomi residential proxy (for rate-limited sources)\nEVOMI_USER=\nEVOMI_PASS=\n```\n\n**Zero-config mode**: YouTube captions work without any API keys. You only need `ANALYSIS_API_KEY` for the `-a` analysis flag and for audio transcription of videos without captions.\n\n## Usage\n\n```bash\n# Transcript only (free for YouTube with captions)\nva \"https://www.youtube.com/watch?v=...\"\n\n# With timestamps\nva \"https://www.youtube.com/watch?v=...\" -t\n\n# AI analysis\nva \"https://www.youtube.com/watch?v=...\" -a\n\n# Analysis + verification (double-pass, uses prompt cache)\nva \"https://www.youtube.com/watch?v=...\" -a -v\n\n# Custom prompt\nva \"https://www.youtube.com/watch?v=...\" -q \"list every product mentioned\"\n\n# Custom prompt + verification\nva \"https://www.youtube.com/watch?v=...\" -q \"extract all claims\" -v\n\n# JSON output (pipe-friendly)\nva \"https://www.youtube.com/watch?v=...\" -a -j | jq .analysis.summary\n\n# Non-YouTube (requires ANALYSIS_API_KEY)\nva \"https://x.com/user/status/123456789\"\nva \"https://www.tiktok.com/@user/video/123456789\"\n\n# Force audio transcription (skip captions)\nva \"https://www.youtube.com/watch?v=...\" -f\n\n# Different language\nva \"https://www.youtube.com/watch?v=...\" -l ro\n\n# Open cache folder\nva -o\n\n# Skip cache\nva \"https://www.youtube.com/watch?v=...\" --no-cache\n```\n\n## Flags\n\n| Flag | Short | Description |\n|------|-------|-------------|\n| `--analyze` | `-a` | Analyze transcript with LLM |\n| `--verify` | `-v` | Double-pass verification (implies `-a`) |\n| `--prompt` | `-q` | Custom analysis prompt (implies `-a`) |\n| `--json-output` | `-j` | JSON output to stdout |\n| `--timestamps` | `-t` | Show timestamps in text output |\n| `--force-transcribe` | `-f` | Skip captions, use audio transcription |\n| `--lang` | `-l` | Caption language codes (default: en) |\n| `--no-metadata` | `-M` | Skip metadata fetching |\n| `--proxy` | `-p` | Use Evomi residential proxy |\n| `--model` | | Transcription model (default: `google/gemini-3.1-flash-lite`) |\n| `--analysis-model` | | Analysis model (default: from env) |\n| `--no-cache` | | Bypass all caching |\n| `--open-cache` | `-o` | Open cache folder in file manager |\n\n## Cache\n\nTranscripts, metadata, and analysis results are cached at `~/.cache/video-analyzer/` with human-readable filenames:\n\n```\nAI-Engineer_MCP-UI-Extending-the-frontier_o-zkvb0iFDQ.transcript.json\nAI-Engineer_MCP-UI-Extending-the-frontier_o-zkvb0iFDQ.analysis.gemini_gemini-3.1-flash-lite-preview.json\nAI-Engineer_MCP-UI-Extending-the-frontier_o-zkvb0iFDQ.metadata.json\n```\n\nAnalysis is cached per model — switching `--analysis-model` triggers a fresh analysis. Use `--no-cache` to bypass, or `va -o` to browse cached files.\n\n## Cost\n\n| Operation | Cost |\n|-----------|------|\n| YouTube captions | Free |\n| Audio transcription (Gemini 3.1 Flash Lite) | ~$0.001-0.002/min |\n| Analysis (depends on model) | ~$0.001-0.01/run |\n| Verification pass | ~33% of analysis (prompt cache) |\n| Cached results | Free |\n\n## License\n\nMIT\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftesty-cool%2Fvideo-analyzer-ai","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ftesty-cool%2Fvideo-analyzer-ai","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftesty-cool%2Fvideo-analyzer-ai/lists"}