OpenCues brings AI to the text field you are already typing in. Type _ where you want help and the model fills it in place: answers, calculations, translations, rewrites, live system values. No separate app, no copy-paste round trip, no submit button.
Where it works
OpenCues runs today in Chrome (regular inputs and rich editors), Claude Code, OpenCode, Gemini CLI, and any terminal via the standalone oc-shell wrapper. The same config files work in every host, so a cue or blank you set up once follows you everywhere.
What you can do without leaving the input
| You type | You get |
|---|---|
hey can u send me that report when u get a sec make this formal _ |
Could you please send me that report at your earliest convenience? |
4 + 4 = _ |
4 + 4 = 8 |
hello world translate to japanese _ |
こんにちは世界 |
draft an email to my landlord asking for a rent reduction _ |
the email, written |
ffmpeg command to convert a video to web-ready mp4 _ |
ffmpeg -i input.mov -c:v libx264 -preset slow -crf 23 -c:a aac -b:a 128k -movflags +faststart output.mp4 |
And beyond generation: read and change live system state (volume _, brightness _, weather tokyo _), see word alternatives inline as you compose and cycle them in place (Ctrl+Alt+←/→ to select a word, Ctrl+Alt+↑/↓ to cycle), or run a continuous editing task (agentically correct spelling _).
Bring your own model
OpenCues is model agnostic: plug in Cerebras (recommended), Groq, OpenAI, Anthropic, Gemini, OpenRouter, or run local models via Ollama. The software is free and open source, and the file formats are an open standard anyone can implement.
Start with the repo README, or read what a blank is first.
Related: GitHub repo