Scheduled background tasks
Runs on the user's machine with a background daemon that supports scheduled routines and continues tasks when the app is closed.
Loading page…
Coding / product dossier
Open-source local agent with scheduled tasks, crash recovery, editable Markdown memory and local-model support
Product brief
Bits is a personal agent that runs on your own machine and never sleeps. A small background daemon does the work: routines wake it on a schedule, and tasks keep going when the app is closed. Every step is saved, so a crash just resumes. Memory lives in plain Markdown files you can read and edit. Risky actions pass the Sentinel, a reviewer that only trusts your own words. It answers with live cards and mini-apps, talks by voice, opens with ⌥Space, and works with any model, including local ones.
Why we selected it
The strongest differentiated personal-agent option: open source, runs locally, resumes interrupted work and stores editable Markdown memory. Support for local models and a reviewer for risky actions adds technical
Product preview saved with our daily selection
Capability scan
Only capabilities supported by the product information we collected are listed here.
Runs on the user's machine with a background daemon that supports scheduled routines and continues tasks when the app is closed.
Saves each step so work can resume after a crash.
Stores memory in plain Markdown files that users can read and edit.
Passes risky actions through Sentinel, a reviewer described as trusting only the user's own words.
Supports models including local ones, voice interaction, and responses through live cards and mini-apps.
Best-fit use cases
FAQ
Bits is an open-source personal agent that runs on the user's own machine.
The background daemon keeps tasks going when the app is closed. Scheduled routines can also wake the agent.
Bits saves each step so interrupted work can resume after a crash.
Yes. Its memory lives in plain Markdown files that users can read and edit.
Risky actions pass through Sentinel, a reviewer described as trusting only the user's own words.