Personal-use • macOS · Windows
Real-time notes and contextual answers in a minimalist overlay. Captures both sides of the conversation locally, hides cleanly from your screen-share, and pulls context from the job description and resume you paste in.
Personal-use only. You are responsible for compliance with local recording laws and the terms of any meeting platform you use it with. Read the license.
Live transcription, channel-aware diarization, and a hotkey-driven UX that keeps your screen-share clean.
Captures your microphone and the meeting audio as separate channels — you get free speaker attribution without an ML model in the loop.
Paste a job description and your resume once at session start. They stay pinned in the prompt cache so every answer reads the room.
Listen for passive notes. Ask for a single question on demand. Auto for real-time question detection while the other side is talking.
Uses documented OS APIs (NSWindow.sharingType on macOS, WDA_EXCLUDEFROMCAPTURE on Windows) so the overlay does not appear in Zoom, Meet, or Teams shared video.
Audio is processed locally before transcription. No telemetry, no analytics, no cloud accounts. Your transcripts are not persisted.
Claude Sonnet 4.6 streaming with HuggingFace Mistral-7B as automatic fallback if Anthropic is rate-limited or down. You always get an answer.
Download the .dmg or .msi for your platform. First-run consent flow surfaces the personal-use license and asks for the audio permissions cue needs.
Drop in the job description and your resume. cue pins them in the Anthropic prompt cache so every subsequent answer is contextually grounded.
Cmd/Ctrl + \ shows or hides the overlay. Pick Listen for notes only, Ask for a single targeted question, or Auto for real-time question detection.
When you share your screen, the overlay is excluded from the captured video by the OS itself. You see it. Zoom does not.
One download for your platform. No account. No subscription. Personal use only.
Pick your platform