CodexPotter continuously reconciles code base toward your instructed state (Ralph Wiggum pattern):
- 🤖 Codex-first — Codex subscription is all you need; no extra LLM needed.
- 🧭 Auto-review / reconcile — Review and polish multi rounds until fully aligned with your instruction.
- 🚀 Never worse than Codex — Drive Codex, nothing more; no business prompts which may not suit you.
- 🧩 Seamless integration — AGENTS.md, skills & MCPs just work™ ; opt in to improve plan / review.
- 🧠 File system as memory — Store instructions in files to resist compaction and preserve all details.
- 🪶 Tiny footprint — Use <1k tokens, ensuring LLM context fully serves your business logic.
- 📚 Built-in knowledge base — Keep a local KB as index so Codex learns project fast in clean contexts.
1. Prerequisites: ensure you have codex CLI locally. CodexPotter drives your local codex to perform tasks.
2. Install CodexPotter via npm or bun:
# Install via npm
npm install -g codex-potter# Install via bun
bun install -g codex-potter3. Run: Start CodexPotter in your project directory, just like Codex:
# --yolo is recommended to be fully autonomous
codex-potter --yoloReplay history for an existing .codexpotter project and optionally continue iterating:
# Open the resume picker (no path)
codex-potter resume
# Or resume directly by project path
codex-potter resume 2026/02/01/1
# Continue in yolo mode (optional)
codex-potter resume 2026/02/01/1 --yoloIn the picker: enter to resume, esc to start new, ctrl + c to quit.
- Skill popup
- Resume (history replay + continue iterating)
- Project management
- Better sandbox support
- Interoperability with codex CLI sessions (for follow-up prompts)
- Allow opting out knowledge base
- Recommended skills for PRD and code review
- Better handling of stream disconnect / similar network issues
# Formatting
cargo fmt
# Lints
cargo clippy
# Tests
cargo nextest run
# Build
cargo build
This project is community-driven fork of openai/codex repository, licensed under the same Apache-2.0 License.
