Package Information
Released: 3 mo ago
Downloads: 0 weekly / 0 monthly
Latest Version: 0.1.0
Author: Hüseyin Tıntaş
Available Nodes
Documentation generating…
Documentation
n8n-nodes-klaket
n8n community node for Klaket — turn any video into LLM-ready data inside your workflows: word-timestamped transcripts (~100 languages), speaker labels, scenes + keyframes, on-screen text (OCR), chapters, SRT/VTT subtitles and in-video moment search. Fully local, no API keys.
Installation
In n8n: Settings → Community Nodes → Install → n8n-nodes-klaket
You also need a running Klaket API (self-hosted, free):
git clone https://github.com/huseyinstif/klaket.git && cd klaket
docker compose up -d --build # API on :8484
n8n running in Docker? Set the credential's API URL to
http://host.docker.internal:8484(or your host's address) instead oflocalhost.
Credentials
Klaket API — just the API URL (default http://localhost:8484). API key only if you enabled KLAKET_AUTH=on.
Operations
| Operation | What it does |
|---|---|
| Ingest & Wait | Queue a video and wait for the full LLM-ready result (transcript, scenes, chapters…) |
| Ingest | Queue a video, return the job ID immediately (pair with a webhook) |
| Get Job Status | Progress of a job (stage + percent) |
| Get Result | Result of a completed job |
| Find Moment | Search inside a processed video — diacritic-insensitive and typo-tolerant |
Example workflow ideas
- Meeting pipeline: Webhook (recording URL) → Klaket Ingest & Wait → LLM summarize → post to Slack
- YouTube watchlist: RSS trigger → Klaket → store transcript chunks in a vector DB
- Podcast subtitles: Dropbox trigger → Klaket → email the SRT to your editor
MIT licensed.