Aikito
A CLI that keeps coding agents' shared context in one place: instructions, skills, MCPs, subagents and durable memory across projects and machines
Medium risk
We rate an entry medium when the tool runs code, makes network calls or reads project files. Check what exactly it does before installing.
Why this level
- It writes and moves agent configuration across directories
- With flags it can delete and overwrite existing settings
Install
Manual install
uv tool install aikitoAfter install: aikito init workspace ~/aikito, then adopt and sync.
This is third-party code. Review the repository files before installing.
What it does
Aikito gathers into one workspace the settings that coding agents share. Through it you manage instructions, skills, MCP servers, subagents and durable memory for several agents, projects and machines at once. The adopt command pulls in an agent's existing setup, sync lays it back into the right places, and status shows the current state. Before writing, adopt and sync first compute a full plan, so you can see what will change.
Who it is for. For developers who use several coding agents and want their settings kept consistent.
Good fit when
- You use several agents and want one set of instructions, skills and MCPs
- You need to move an agent's setup between projects and machines
- You want shared durable memory and subagents across agents
Not a fit when
- You use a single agent and a single project
- You are not ready for a tool that lays config into agent directories
- Manual setup without a shared workspace is enough
Example request
Set up an Aikito workspace, adopt my current agent setup and show the sync planLimitations
Aikito writes and moves agent configuration, so review the plan before applying; adopt and sync show it. It needs Python 3.12 or newer and installs via uv or Homebrew. Windows requires developer mode. Documentation is partly in English and Chinese. MIT licensed.
How to disable. Uninstall the tool installed via uv (uv tool uninstall aikito) or Homebrew and, if needed, remove the created workspace.
Security check
- It writes and moves agent configuration across directories
- With flags it can delete and overwrite existing settings
README in short
The README describes Aikito as one place to manage coding agents' shared context: instructions, skills, MCPs, subagents and durable memory across projects and machines. Install goes through uv or Homebrew, then workspace init and the adopt, sync and status commands. It stresses that adopt and sync compute a plan before writing, and that risky operations need explicit flags. The tool is written in Python and claims to run on several operating systems with no external dependencies. MIT licensed.
FAQ
Does Aikito change my files right away?
No, adopt and sync first compute a full plan, and with --dry-run you can preview changes before writing.
What can the workspace hold?
Instructions, skills, MCP servers, subagents and durable memory, shared across agents, projects and machines.
Related
Open-source personal AI assistant on your own machine: answers in Telegram, Slack, Discord and WhatsApp, extended with skills and plugins
A self-improving agent from Nous Research with a TUI, messaging gateway, cron jobs and skills it writes itself
An open source coding agent for the terminal and desktop with build and plan modes
Open prompt library with a Claude Code plugin, MCP server and CLI: search, fetch and improve prompts and skills from an agent