Turn a YouTube transcript into ChatGPT-ready context.
Clean a YouTube transcript, estimate tokens, split long captions into safe parts, and copy a grounded context pack for ChatGPT. Free, local, and no account.
The pack separates instructions from untrusted transcript text, preserves available timestamps, and splits long material into predictable parts.
No API key required The page prepares text locally. You decide where to paste it and which AI account to use. Grounded by default The generated instructions ask the model to use supported claims, preserve timestamps, and say when the transcript lacks an answer.
Workflow steps: Copy captions from YouTube's Show transcript panel. Paste them here and choose the maximum tokens per part. Copy the context pack into ChatGPT or download Markdown.
Common questions: Does this send the transcript to ChatGPT? No. It creates a context pack locally and copies it only when you press the button. Can it handle a long transcript? Yes. It estimates tokens and divides the text into parts using the limit you choose. Is the token count exact? No. It is a transparent character-based estimate; exact counts vary by model and tokenizer.
The selected media file is processed locally in the browser and is not uploaded to Kamnevapps.