How do I install the Migrate Claude skill?

Run npx skillmds@latest add frabcd/migrate-claude in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Migrate Claude safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Migrate Claude?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Migrate Claude free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Migrate Claude?

frabcd (@frabcd) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Migrate Claude by frabcd — Agent Skill

Name: Migrate Claude
Author: frabcd

Migrate Claude

Inspect a Claude Code game project and plan a reversible Codex migration of guidance, skills, agents, hooks, and paths. Use when CLAUDE.md or .claude content is present; preserve the source and require digest confirmation before writing Codex files.

frabcd 4b216ab 2 files · 1.8 KB Updated Sep 17, 2026

File contents

Migrate a Claude project

Inspect CLAUDE.md, .claude/, existing AGENTS.md, .codex/, engine markers, and uncommitted work. Do not modify them.
Run the core migrate claude --project <root> --output <plan.json> command.
Review every proposed destination. Claude slash commands become namespaced $ai-game-studio:<skill> references; engine/project guidance becomes root or scoped AGENTS.md; supported agent definitions belong under .codex/agents/.
Block unsupported Claude-only model names, tool declarations, memory fields, fixed turn limits, or paths from entering runtime files. Keep the original Claude files unchanged for comparison and rollback.
Present exact actions, detected environment, backups, license notes, permissions, expiry, rollback operations, and digest. Wait for verbatim digest confirmation.
Apply through pack apply --project <root> --plan <plan.json> --confirmed-digest <digest>.
Run validate skills, validate platform, project tests, and a representative Codex task. Report the transaction ID and how to generate a rollback plan.

If destinations already contain user-authored guidance, stop at the proposal and show the overlap. Never silently merge policy text whose precedence is unclear.

Frequently asked questions