# Unity AI CLI Bridge

> Interact with a local Unity Editor through `unity-ai-cli`. Use when you need to: (1) Discover available Unity tools dynamically, (2) Execute `unity-ai-cli` tools with JSON arguments, (3) Expose the Unity Editor as an MCP server for AI clients.

- Skill: `hatayama/unity-ai-cli-bridge` (Agent Skill)
- Install (CLI): `npx skillmds@latest add hatayama/unity-ai-cli-bridge`
- Raw SKILL.md: https://api.skillmd.com/api/skills/hatayama/unity-ai-cli-bridge/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: hatayama (https://skillmd.com/u/hatayama)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/hatayama/unity-ai-cli-bridge

---


# Unity AI CLI Bridge

Use this skill when you need to interact with the local Unity Editor through `unity-ai-cli`.

## Workflow

1. Confirm the Unity project is open.
2. Use `unity-ai-cli tools list --json` to discover the currently available tools.
3. Use `unity-ai-cli tools describe <tool>` before calling a tool you have not used yet.
4. Use `unity-ai-cli tools call <tool> --json-args '<json>'` for direct human-driven execution.
5. Use `unity-ai-cli serve-mcp` when an AI client needs standard MCP `tools/list` and `tools/call`.

## Important Rules

- Tool discovery is dynamic. Do not assume a static tool catalog.
- Prefer discovery before execution when the project or package state may have changed.
- Dangerous editor mutations should still be confirmed explicitly in the conversation.
- If Unity reports connection approval pending, accept the client in Project Settings > AI > Unity MCP before retrying.

## Common Starting Points

- Read console output:
  - `unity-ai-cli tools call Unity.ReadConsole --json-args '{}'`
- Inspect available tools:
  - `unity-ai-cli tools list --json`
- Expose Unity to an MCP client:
  - `unity-ai-cli serve-mcp`