AI Agents

TypeSafe's decision model: it returns fast, structured decisions that software can use directly, instead of writing text. In early access through an API.

Anthropic's coding agent, in the terminal or your editor

Self-hosted assistant with memory, reachable from Telegram or Slack

Open-source shared memory for AI assistants, stored as Notion pages, from Notion. Claude Code, Codex, Cursor and other MCP hosts read and write it. MIT licensed.

Build a company agent that works inside Slack, GitHub and Linear

A browser your coding agent can drive from the terminal

Coding agent in a single binary, no runtime to install

Coding agent built for speed, in the terminal or on the desktop

Self-hosted workspace where agents work from your company files

Lets your agents send and receive email, WhatsApp and SMS

Open-source agent workspace built on your company's own data

Agent framework from DeepSeek where every part is a plugin

Framework for agents that run on Vercel or your own servers

Open-source memory for agents, stored as a graph

Block's open-source coding agent, works with any model

A minimal coding agent for the terminal, written in C

Tiny open-source agent built around bash

Shrinks logs and tool output before the model sees them

Hosts your AI agents on tiny VMs that sleep when idle

Shared memory your agents can read and write across sessions

Build and run AI agents without writing code

Composes and governs agents on top of Claude Code, Codex and Cursor

Coding agents in isolated worktrees, with diffs reviewed before PRs

Open-source assistant on your machine, answers in WhatsApp or Slack

Run a fleet of coding agents in parallel on your own subscription

Run a team of AI agents with an org chart and budgets

Desktop workspace for coding agents, one Git worktree per task