EvilFreelancer

Rpa Gen Rules — Development skill for Claude Code

Development community

Agent skill for Claude Code, Codex and Cursor: generate or refresh project rules (.cursor/rules/.mdc, CLAUDE.md and .claude/rules, a .codex hook bridge that attaches Cursor rules by glob) with layere.

How to install Rpa Gen Rules

This entry records only its repository, not the path inside it, so there is no exact command to give. Open EvilFreelancer/rpa-gen-rules and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Rpa Gen Rules does

Agent skill for Claude Code, Codex and Cursor: generate or refresh project rules (.cursor/rules/*.mdc, CLAUDE.md and .claude/rules, a .codex hook bridge that attaches Cursor rules by glob) with layered-cake architecture, BDD delivery and a mandatory Rules Sync step across every agent tree

Alternatives in Development

Lamda — The most powerful Android RPA agent framework, next generation mobile automation 7.7k ★
Agent Rules — Rules and Knowledge to work better with agents such as Claude Code or Cursor 5.7k ★
Cyberboss — 一个接入微信的本地生活 Agent Bridge，让 Codex / Claude Code 拥有时间感、行踪感、随机唤醒和自主唤醒能力，用主动陪伴替代所有番茄钟和效率工具，自动记录日记、维护生活时间轴、发送文件和表情包 1.3k ★

README

Agent project rules

Purpose

Create or update **agent rules** for **Cursor** (`.cursor/rules/*.mdc`), **Claude Code** (`.claude/` with modular `rules/`), and **OpenAI Codex** (a `.codex/` hook bridge that auto-attaches Cursor rules by glob, since Codex has no native equivalent). Bundled examples live under `references/`.

Rules follow a **layered-cake** idea: implement **inner layers first** (no internal project dependencies), then the next layer, and so on. They also describe **BDD-style** work (behavior driven by tests). No long user brief is required; the agent infers from the repo.

Generated rules, `AGENTS.md`, and every other agent brief are written **in English** unless you explicitly ask for another language - mirrored trees are only reviewable when both sides speak the same language. The chat itself stays in your language.

Generated rules **must** include a mandatory **Rules Sync** step: any change to a rule for one agent (for example a Cursor `.mdc`) has to be mirrored to every other agent's tree (Claude `.md`, root `AGENTS.md` / `CLAUDE.md`, the `.codex/rules.md` index, etc.) in the **same** commit. The skill also recommends keeping `CLAUDE.md` as a symlink to `AGENTS.md` at repo root (see `getconf`, `getconf-ui`, `sdm-client-proxy` for reference). See `SKILL.md` and `references/bdd-and-agents.md` for the exact wording and translation table.

For Codex the skill installs a ready-made hook bridge (`.codex/hooks.json` + `.codex/hooks/attach_rules.py`, copied verbatim): `alwaysApply: true` rules are injected at `SessionStart`, glob-matched rules on `PreToolUse` when a patch touches covered files - the same behavior Cursor and Claude Code provide natively. Rule bodies are **not** duplicated; the hook reads `.cursor/rules/*.mdc` directly. After install, run `/hooks` in Codex once to trust the hook.