Quick Start › Integrations

Integrations

How KaKeKiKoKu Studio connects with the MCP-capable AI clients you already use

KaKeKiKoKu Studio is a local MCP server. To use it you need an AI client that speaks MCP — and that you already subscribe to (KaKeKiKoKu does not consume any AI API; it works through your existing client subscription).

If your client is not listed below, head to Other MCP clients — the generic guide covers patterns that should work with any MCP-compatible client (Cursor, Windsurf, VS Code MCP extensions, etc.).

Two pieces every integration needs

Connecting KaKeKiKoKu Studio to your AI has the same shape regardless of which client you use:

  1. Point your AI to KaKeKiKoKu Studio via MCP — register KaKeKiKoKu as an MCP server in your AI client.
  2. Add a minimal startup instruction — a small file or rule that tells your AI to use KaKeKiKoKu Studio at the start of every session.

The mechanics differ — Claude Code and Claude Desktop's Code tab use .claude/rules/wuniq.md, Codex uses AGENTS.md, and Antigravity 2.0 uses an Always On workspace rule under .agents/rules/, for example — but the goal is identical: your AI knows about KaKeKiKoKu Studio and uses it from the first message. Each guide below covers both pieces for its specific client.

Model-agnostic does not mean every session feels identical. KaKeKiKoKu Studio is not tied to any specific AI model or provider. The real experience can still vary with the model you choose, the client implementation, your plan, available context, and the provider's service health at that moment. If a provider is degraded, overloaded, or a model is simply performing poorly in a given session, KaKeKiKoKu may still be configured correctly even if the interaction feels less smooth.

Supported clients and setup guides

Last reviewed: 2026-07-22. Click any client to open its full setup guide.

Status legend
  • ✓ Validated — tested by the author on a recent version
  • ⚠ Not tested — should work; not officially measured yet
  • ○ Future scope — deferred or blocked by platform roadmap; no active validation
  • ✕ Not available — vendor does not ship this combination
Client Windows macOS Linux
Anthropic
Claude Code Anthropic's CLI — only client with channels support today (Automatic mode). ✓ Validated ○ Future scope ✓ Validated
Claude Desktop Anthropic's desktop app, Code tab — Precise mode. Linux client currently beta. ✓ Validated ○ Future scope ⚠ Not tested
OpenAI
ChatGPT Desktop (Codex) OpenAI's Windows app with local Codex projects — Precise mode. Shares MCP configuration with Codex CLI and the IDE extension. ✓ Validated ○ Future scope ✕ Not available
Codex CLI OpenAI's terminal client — Precise mode. Requires stderr wrapper for upstream bug (see guide). ⚠ Not tested ○ Future scope ✓ Validated
Google
Antigravity 2.0 Google's standalone command center for agent projects — Precise mode. The guide also preserves IDE and CLI configuration paths. ✓ Validated ○ Future scope ✓ Validated
Multi-model
OpenCode Open-source coding agent for many providers and models — Precise mode. ✓ Validated ○ Future scope ⚠ Not tested
Z.ai
ZCode Z.ai's agentic development environment for GLM — Precise mode. Linux client currently beta. ⚠ Not tested ○ Future scope ⚠ Not tested
Other MCP clients
Generic MCP setup Patterns for other MCP clients, including Kimi Code, Qwen Code, Cursor, Windsurf, and similar tools. ⚠ Not tested per client ○ Future scope ⚠ Not tested per client

Linux is available on the Snap Store. Install it from the Download page, then follow the guide for your AI client. macOS is on the roadmap. KaKeKiKoKu Studio does not currently ship a macOS build, so no client can be validated on macOS today — not because of the client, but because KaKeKiKoKu itself is not there yet. Planned for the future, no ETA.

Beyond the dedicated guides

The MCP ecosystem moves faster than this matrix. Clients without a dedicated annex may still work well — we just have not measured them, and we do not commit to documenting every client-specific bug.

Found a KaKeKiKoKu Studio + client combination that works well and is not in the matrix? Tell us — we may add it to a future review.

Need the basics first?

Quick Start guide →   ·   Full documentation →