What it does

Pieces describes itself as a memory layer for everything you do. It runs in the background on your computer and forms memories every two seconds from the apps you already use, building a chronological, searchable timeline of research, chats, emails, notes and meetings. You can ask about anything you have done, scoping questions by time, source or topic, and generate items such as standup updates, morning briefs, meeting prep and day recaps from your real context. The page lists sources including Gmail, Outlook, Chrome, Slack, Zoom, Notion, Figma, Cursor, VS Code and JetBrains, as well as AI tools such as Claude and ChatGPT. Memories are saved on-device by default. You can pause Pieces, disable capture from specific apps and websites, and delete data by time, app, website or capture method. The Pieces MCP Server carries that history into Claude, Cursor, Codex and other MCP-ready assistants, and teams and enterprise offerings add shared knowledge, bring-your-own API keys and org-wide capture policies.

Capabilities

Background capture into a chronological, searchable timeline
Ask questions scoped by time, source or topic
Generates standup updates, morning briefs, meeting prep and day recaps
On-device storage by default, with pause, per-app controls and deletion
MCP server that shares your history with Claude, Cursor, Codex and other assistants
Team and enterprise offerings with org-wide capture policies

When to use it

You lose track of research trails spread across many tabs and apps
You want recaps or meeting prep built from what you actually did
You want your AI assistants to start with context from your earlier work