How do I install the Repo Clone skill?

Run npx skillmds@latest add george-rd/repo-clone 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 Repo Clone 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 Repo Clone?

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 Repo Clone free to use?

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

Who published Repo Clone?

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

Repo Clone by george-rd — Agent Skill

Name: Repo Clone
Author: george-rd

Repo Clone

Use when porting code between languages, migrating a codebase, translating source code, or rewriting in a new language. Trigger phrases: "port", "clone to", "migrate to", "translate to", "rewrite in", "convert from X to Y".

george-rd 54419b8 2.0 KB Updated Sep 17, 2026

File contents

Codebase Porting

Port codebases between languages using manifest-driven headless loops with citation-backed behavioral specs.

Quick Start

Run /repo-clone init <source-lang> <target-lang> to scaffold the porting project
Review the generated porting/manifest.json — it tracks every file through each phase
Run the headless extraction and porting loops:
- while :; do cat PROMPT_extract_tests.md | claude -p --model haiku --dangerously-skip-permissions; sleep 5; done
- while :; do cat PROMPT_extract_src.md | claude -p --model haiku --dangerously-skip-permissions; sleep 5; done
- while :; do cat PROMPT_port.md | claude -p --model haiku --dangerously-skip-permissions; sleep 5; done

How It Works

Init: Scans source repo, categorizes files (test/source/config/asset/doc), builds porting/manifest.json with per-file tracking
Phase 1 (extract-tests): Loop extracts behavioral specs from test files with [test:file:line-range] citations
Phase 2 (extract-src): Loop extracts behavioral specs from source files with [source:file:line-range] citations, cross-referencing test specs
Phase 3 (plan): Synthesize IMPLEMENTATION_PLAN.md from all specs
Phase 4 (build): Loop ports one task per iteration, following citations to original source
Phase 5 (audit): Parity check between source and ported code

Each loop iteration reads the manifest, finds the next pending file, processes it, and marks it done. All state lives on disk -- no memory between iterations.

Full methodology: references/methodology.md Prompt templates: data/templates/PROMPT_extract_tests.md, PROMPT_extract_src.md, PROMPT_port.md Cross-language patterns: references/semantic-mappings.md

---name: repo-clonedescription: Use when porting code between languages, migrating a codebase, translating source code, or rewriting in a new language. Trigger phrases: "port", "clone to", "migrate to", "translate to", "rewrite in", "convert from X to Y".---# Codebase PortingPort codebases between languages using manifest-driven headless loops with citation-backed behavioral specs.## Quick Start1. Run `/repo-clone init <source-lang> <target-lang>` to scaffold the porting project2. Review the generated `porting/manifest.json` — it tracks every file through each phase3. Run the headless extraction and porting loops: - `while :; do cat PROMPT_extract_tests.md | claude -p --model haiku --dangerously-skip-permissions; sleep 5; done` - `while :; do cat PROMPT_extract_src.md | claude -p --model haiku --dangerously-skip-permissions; sleep 5; done` - `while :; do cat PROMPT_port.md | claude -p --model haiku --dangerously-skip-permissions; sleep 5; done`## How It Works- **Init**: Scans source repo, categorizes files (test/source/config/asset/doc), builds `porting/manifest.json` with per-file tracking- **Phase 1 (extract-tests)**: Loop extracts behavioral specs from test files with `[test:file:line-range]` citations- **Phase 2 (extract-src)**: Loop extracts behavioral specs from source files with `[source:file:line-range]` citations, cross-referencing test specs- **Phase 3 (plan)**: Synthesize IMPLEMENTATION_PLAN.md from all specs- **Phase 4 (build)**: Loop ports one task per iteration, following citations to original source- **Phase 5 (audit)**: Parity check between source and ported codeEach loop iteration reads the manifest, finds the next pending file, processes it, and marks it done. All state lives on disk -- no memory between iterations.Full methodology: `references/methodology.md`Prompt templates: `data/templates/PROMPT_extract_tests.md`, `PROMPT_extract_src.md`, `PROMPT_port.md`Cross-language patterns: `references/semantic-mappings.md`

Frequently asked questions