Cc clone

Skill duliangkuan/cc-clone/skill/cc-clone

Clone a Claude Code brain — package/redact/distribute memory, skills, agents & rules; one-click install others' brains; export to Codex/Gemini/Cursor/OpenCode.

Install

npx -y skills add duliangkuan/cc-clone --skill cc-clone

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

22 days oldThe repository was created 22 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Clone a Claude Code brain — package your own memory/skills/agents/rules into a shareable, secret-redacted pack, or one-click install another developer's brain into your Claude Code (or export it to Codex/Gemini/Cursor/OpenCode). Use when the user says "clone my brain", "package my Claude Code setup", "install someone's brain", "克隆大脑", "打包我的记忆", "移植到 codex/cursor", or wants to distribute or adopt a Claude Code configuration.

SKILL.md

3.2 KB, as published. Nobody here has run it

cc-clone

Turn a Claude Code "brain" (its CLAUDE.md, skills, subagents, commands, rules) into a portable pack that anyone can install in one step — with secrets stripped automatically.

When to use

"Package / clone my brain" → run pack, then show the redaction report.
"Install X's brain" / a gh: or repo URL → run install.
"Move my brain to Codex / Cursor / Gemini / OpenCode" → run install --to <host>.
"What brains do I have?" → run list.

Prerequisite

cc-clone is a small Python (stdlib-only) CLI. If cc-clone is not on PATH, run it from a clone of the repo with python -m cc_clone …, or pip install cc-clone. Check with cc-clone doctor.

Recipes

Package the user's brain for sharing

cc-clone pack <brain-name> --author "<name>" --description "<one line>"

Defaults to the global brain (~/.claude) and the reusable components (skills, agents, commands, rules, CLAUDE.md). Personal auto-memory and MCP configs are excluded by default — add them only on request with --include memory,mcp (they are the riskiest for leaks).

ALWAYS show the user the 🔒 redaction report and the pack path, and tell them to eyeball the pack before publishing. Never git push a brain the user hasn't reviewed. Secrets are masked, but PII (emails/IPs) is only flagged unless --strict is passed.

Install another developer's brain

cc-clone install gh:someuser/their-brain     # GitHub repo
cc-clone install ./path/to/pack              # local pack
cc-clone install <url> --dry-run             # preview first

Installs are non-destructive: skills/agents are namespaced (<brain>__<name>), CLAUDE.md content is fenced, and every install writes a receipt so cc-clone uninstall <name> fully reverses it. Remind the user to restart Claude Code afterward.

Move a brain to another agent

cc-clone install gh:someuser/their-brain --to codex     # or gemini | cursor | opencode

CLAUDE.md becomes AGENTS.md (or GEMINI.md), skills/agents/MCP are translated to the target's native format. Some artifacts have no equivalent on some hosts — relay the tool's ⚠️ notes to the user honestly, don't pretend it was 1:1.

Verify integrity before trusting a pack

cc-clone verify ./path/to/pack

Safety rules (do not skip)

Before publishing a pack, surface the full redaction report to the user.
Never auto-git push. Packing and publishing are separate, deliberate steps.
When installing from a URL, run verify (install does this automatically) and report any checksum mismatch as a red flag.
MCP servers are never silently merged on install — they need the user's own keys. Point the user at the pack's mcp.json.