Official description7 sourcesMCP servers8d ago
huseyinstif/klaket
Local video understanding: word-timest
View on GitHub ↗Direct link · no redirect page
Config template (adjust for your client)
claude mcp add klaket -- npx -y klaket
A pasteable skeleton generated from the repository name. Check the project README for the exact package and arguments.
- Star count
- 3 ★ · huseyinstif/klaket · read from the GitHub search API · (checked 2026-10-16 20:00:20)
- Source
- collected from the punkpeye/awesome-mcp-servers list · type: MCP server
- Harvest history
- first seen 2026-10-09 · last seen 2026-10-16 · seen 32× · listed by 7 sources
- Score
- score 703 · 3 ★ · 7 sources · mcp
FAQ (answers come from our records)
How do I install it?
We did not find an install command for this repository (our install coverage is only about 4%). Check its README or releases page.
Who is it for?
We filed it under these tags: MCP 服务. The tags come from the entry itself, not a taxonomy we imposed.
Who maintains it?
repository
huseyinstif/klaket；first catalogued 2026-10-09；last seen 2026-10-16。There is no official certification or quality rating here — we record sources and dates only.Third-party quality signals
🎬 Klaket
以下摘录自该仓库的 README（片段） · 在 GitHub 读完整 README →
**Turn any video into LLM-ready data.**
> A *klaket* is a clapperboard — the tool that syncs sound and image on a film set. **Klaket syncs video with LLMs.**
LLMs read text. The web became readable with scrapers — but video, the largest store of human knowledge, is still locked away. Klaket unlocks it: give it a video URL or file, get back structured, timestamped, LLM-ready data.
Features
- **📝 Transcript** — timestamped speech-to-text in **~100 languages** (auto-detected) with **word-level timestamps**; pick the model per job ("model": "medium")
- **🎙️ Podcasts too** — pass an audio file/URL (mp3, m4a…) and Klaket skips the visual stages, deriving chapters from speech pauses
- **🗣️ Speaker diarization** — who said what (S1/S2/…), local & keyless (sherpa-onnx)
- **💬 Subtitles** — ready-to-use .srt / .vtt files with speaker labels
- **🎞️ Scene detection** — content-aware scene boundaries + keyframes per scene
- **🔎 On-screen text (OCR)** — reads slides, terminals and captions per scene, local & keyless
- **🧩 One JSON timeline** — transcript, scenes, frames and on-screen text aligned on a single timeline
- **🔌 Works offline, no API key required** — the core pipeline uses zero LLM calls
- **🧠 Pluggable model layer** — optional scene descriptions via local VLMs (Ollama) or any OpenAI-compatible endpoint (KLAKET_VLM=off by default)
- **🤖 MCP server** — let coding agents "watch" any video and find moments inside it
- **🔍 In-video search** — GET /v1/jobs/{id}/search?q=… finds the exact moment
- **▶️ Playground** — the dashboard plays the video with a click-to-seek, live-highlighted transcript
SDKs
Give your agent eyes
Then: *"Watch https://youtube.com/watch?v=… and summarize the commands the presenter runs."*
The agent gets klaket_ingest, klaket_job_status and klaket_get_result tools.
Quick start
That's it — no API keys, no GPUs required. make help lists developer shortcuts (make up, make test, make e2e).
Architecture
- apps/api — Go, job orchestration
- apps/worker — Python, media pipeline
- apps/dashboard — React dashboard
Self-host vs Cloud
> A *klaket* is a clapperboard — the tool that syncs sound and image on a film set. **Klaket syncs video with LLMs.**
LLMs read text. The web became readable with scrapers — but video, the largest store of human knowledge, is still locked away. Klaket unlocks it: give it a video URL or file, get back structured, timestamped, LLM-ready data.
Features
- **📝 Transcript** — timestamped speech-to-text in **~100 languages** (auto-detected) with **word-level timestamps**; pick the model per job ("model": "medium")
- **🎙️ Podcasts too** — pass an audio file/URL (mp3, m4a…) and Klaket skips the visual stages, deriving chapters from speech pauses
- **🗣️ Speaker diarization** — who said what (S1/S2/…), local & keyless (sherpa-onnx)
- **💬 Subtitles** — ready-to-use .srt / .vtt files with speaker labels
- **🎞️ Scene detection** — content-aware scene boundaries + keyframes per scene
- **🔎 On-screen text (OCR)** — reads slides, terminals and captions per scene, local & keyless
- **🧩 One JSON timeline** — transcript, scenes, frames and on-screen text aligned on a single timeline
- **🔌 Works offline, no API key required** — the core pipeline uses zero LLM calls
- **🧠 Pluggable model layer** — optional scene descriptions via local VLMs (Ollama) or any OpenAI-compatible endpoint (KLAKET_VLM=off by default)
- **🤖 MCP server** — let coding agents "watch" any video and find moments inside it
- **🔍 In-video search** — GET /v1/jobs/{id}/search?q=… finds the exact moment
- **▶️ Playground** — the dashboard plays the video with a click-to-seek, live-highlighted transcript
SDKs
Give your agent eyes
Then: *"Watch https://youtube.com/watch?v=… and summarize the commands the presenter runs."*
The agent gets klaket_ingest, klaket_job_status and klaket_get_result tools.
Quick start
That's it — no API keys, no GPUs required. make help lists developer shortcuts (make up, make test, make e2e).
Architecture
- apps/api — Go, job orchestration
- apps/worker — Python, media pipeline
- apps/dashboard — React dashboard
Self-host vs Cloud
- Source
- awesome · https://github.com/huseyinstif/klaket
- Browse by field
- MCP servers
- Traction
- 3 ★
- added
- yesterday
More in MCP servers
Filed under
Every tag opens the full list for that filter. Counts are the live catalogue figures.