Local-first AI knowledge layer. Extract architecture, query from any AI tool via MCP. Private by architecture.

Experiment task scheduling made easy.

Your infinite canvas workspace for agents, tasks, knowledge, and research. An infinite canvas for Claude Code, Codex, terminals, tasks, and notes.

Delegate tasks to DeepSeek right inside your Claude Code / Codex sessions.

The living ecosystem where AI agents complete tasks through workflow loops, improve through iterative execution, are evaluated by mentor agents or humans in the loop, and turn completed work into reusable work experience and data to improve future agents.

Web tasks that compile. Other browser tools tell your AI agent the click worked. chrome-agent reads the page back after every action and answers in JSON what actually happened. One 3 MB Rust binary, no Node, no cloud.

The evaluation benchmark on MCP servers

[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.

Collective intelligence runtime for AI agents. Knowledge graph + persistent memory.

Official Model Studio CLI（阿里云百炼 CLI）built for AI Agent frameworks, exposing models, search, multimodal, and workflow capabilities as structured tool calls.