Prompt Editor
Write prompts as plain text or ChatML messages. Token counts and context-window alerts update as you type — everything stays in your browser.
All processing for core tools runs locally in your browser. Privacy policy
How to use
- Choose Text/Markdown or ChatML mode in Options.
- Pick a target model so token counts and context alerts match your API.
- Edit freely — drafts autosave to localStorage; Clear draft wipes them.
- Watch the sidebar: amber at ≥70% context, red at ≥90% or over limit.
FAQ
Does my prompt leave the browser?
No. Editing, token counting, and drafts run locally. Your prompts never leave your browser for this tool.
What is ChatML mode?
ChatML mode edits System / User / Assistant message cards and counts tokens with OpenAI-style message framing overhead.
How accurate are token counts?
OpenAI models use Exact js-tiktoken encodings. Other vendors show Estimate (o200k proxy). Send to TokenWorkbox for finer Exact Qwen/GLM counting.