Introduction
What Overshow is, who it is for, and how local capture and search fit together.
Last updated: 2 April 2026
What is Overshow?
Overshow is a privacy-first tool for screen capture, meeting audio, and local search built for knowledge work. It records what appears on your screen and what is said in active meetings, processes that material on your Mac, and stores it in a local database so you can find it quickly later.
Search combines full-text matching with semantic search, so you can ask in your own words and still surface relevant moments. Answers are grounded in what Overshow has actually captured — nothing is invented beyond your own content.
Why people use it
Privacy-first, on-device processing
Your captures stay on your machine. Processing — including transcription, text extraction, and embeddings — runs locally. That keeps sensitive work inside your environment and reduces dependence on cloud services for day-to-day recall.
Screen capture and meeting audio
In the public beta, Overshow is meeting and audio first: it records audio only while meetings are active or manually started, and speech-to-text turns that audio into searchable transcripts. The public beta does not capture your screen, apps, or documents. Meeting audio is the primary source, and — with your permission — Overshow can also draw on your calendar and connected email accounts for context. (Overshow has on-device screen OCR too, but that is not part of the meeting-focused public beta.) Together these build a searchable record of what was discussed and decided.
AI-powered search with grounded answers
You can search across transcripts and any other enabled local sources using both keyword and meaning-based retrieval. When you use Ask, Overshow finds and presents information drawn from your captures rather than producing generic text from a model alone.
Who it is for
Overshow suits knowledge workers who need reliable recall across tools and conversations: engineers revisiting decisions, operations staff tracing incidents, support teams checking past cases, and product people reconstructing context. It is aimed at individuals and teams who value local control and auditability.
Quick links
- Quickstart — get running in a few minutes
- Installation — download and first launch
- Onboarding — wizard steps explained
- System requirements — hardware and OS
- Permissions — what each permission enables and how to change it