Rule Architect — AI skill for Claude Code
Generate or update a runtime-neutral AI rule set (AI_RULES.md + docs/.md).
How to install Rule Architect
Installs to ~/.claude/skills/moveju112-rule-architect-rule-architect/SKILL.md
mkdir -p ~/.claude/skills/moveju112-rule-architect-rule-architect && curl -fsSL https://raw.githubusercontent.com/moveju112/rule-architect/HEAD/commands/rule-architect.md -o ~/.claude/skills/moveju112-rule-architect-rule-architect/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Rule Architect does
description: Generate or update a runtime-neutral AI rule set (AI_RULES.md + docs/*.md) argument-hint: "[project-path] [--update]"
Invoke the `rule-architect` skill.
Target: $ARGUMENTS (default: current working directory).
Follow the skill's SKILL.md exactly:
scan.pyfor the signal manifest — it decides the conditional doc set, and itstruncatedflags say whether the scan was complete. Stop ifbrokenRuleLinksis non-empty.- On an existing rule set, `manifest.py check
Alternatives in AI
- Pi Agent VS OpenCode — Customization & Control Comparison — Pi v0.52+ vs OpenCode v1.1+ (Feb 2026) Thesis: Pi and OpenCode are both MIT-licensed, open-source, model-agnos 526 ★
- Devo — Model-neutral agent desktop/runtime for private, enterprise, and OpenAI-compatible / Anthropic-compatible mode 316 ★
- Nativeprompt — Rewrite your prompt into the native dialect of the model you're actually running (Claude Code / Codex) — by ea 113 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Continuum
Provider-neutral continuity runtime for AI coding agents — checkpoint, handoff and resume tasks across Codex,
Concept Update
Sweep recent changes and update the LLM-Wiki repo's concept docs, architecture diagram, AGENTS.md, and memory.
Stillcurrent
Stillcurrent keeps the architecture docs and diagrams of your software up to date: LLM agents write and update
AI Workspace Archive
A massive, self-updating local archive of AI tools — 11,000+ agent skills, 240+ MCP servers, 2,200+ IDE rules
Rules As Skills
Encode MUST/NEVER constraints as installable, cross-platform skills for AI agents. Three-layer model gives dyn
Agent Rule Gaming
Experiments in a small RL environment: do LLM agents game the rules when the answer is within reach, and can y
Related Agents
Remotion Rule Reviewer
Remotion 渲染前的"自信关"——在跑贵渲染之前,对当前改动做一次语义级审查。对照 docs/HARD_RULES.md(框架红线)和 kit/theme.ts(设计 token),抓 hook 抓不到的语义错:s
Architect Rules Reviewer
Rule and convention conformance reviewer for technical plans
Review Judge
Neutral zero-context judge for the adversarial-review skill. Rules ONLY on findings the author disputed — read