RuleSync – ESLint and Babel for AI Agent Rules (Claude.md, .cursor/Rules)

Notifications You must be signed in to change notification settings Fork 0 Star 0 BranchesTags Open more actions menu Folders and files NameName Last commit message Last commit date Latest commit History 2 Commits 2 Commits .cline .cline .cursor/rules .cursor/rules .github .github .rulesync .rulesync src src tests tests .env.example .env.example .gitignore .gitignore .windsurfrules .windsurfrules AGENTS.md AGENTS.md CHANGELOG.md CHANGELOG.md CLAUDE.md CLAUDE.md CODE_OF_CONDUCT.md CODE_OF_CONDUCT.md CONTRIBUTING.md CONTRIBUTING.md LAUNCH.md LAUNCH.md LICENSE LICENSE README.md README.md SECURITY.md SECURITY.md SPECIFICATION.md SPECIFICATION.md package-lock.json package-lock.json package.json package.json tsconfig.json tsconfig.json tsup.config.ts tsup.config.ts Repository files navigation The ESLint & Babel for AI Agent Rules — Compile, lint, optimize, and sync instructions across AGENTS.md, CLAUDE.md, .cursor/rules, and Copilot. 💡 What is RuleSync? RuleSync is an open-source CLI engine, static analyzer, and cross-format compiler for AI agent context instruction rules. It allows developers and engineering teams to write a single canonical source of truth (AGENTS.md or .rulesync/) and automatically compile, lint, optimize, and synchronize instructions across Cursor, Anthropic Claude Code, GitHub Copilot, Cline / RooCode, Windsurf, and OpenCode. ┌───────────────────────────┐ │ Source: AGENTS.md │ │ (Canonical Rules File) │ └─────────────┬─────────────┘ │ ▼ ┌───────────────────────────┐ │ Abstract Rule Tree │ │ (ART) │ └─────────────┬─────────────┘ │ ┌─────────────────────┴─────────────────────┐ ▼ ▼ ┌───────────────────────────┐ ┌───────────────────────────┐ │ Static Linter Engine │ │ Multi-Target Compiler │ │ ├─ Duplicate check │ │ ├─ CLAUDE.md │ │ ├─ Conflict detector │ │ ├─ .cursor/rules/*.mdc │ │ ├─ Token budget check │ │ ├─ copilot-instructions │ │ └─ Security sanitizer │ │ ├─ .cline/instructions │ └───────────────────────────┘ │ └─ .windsurfrules │ └───────────────────────────┘ 🎯 Who is it for? Polyglot AI Developers: Working with 2+ AI tools (e.g. Claude Code CLI + Cursor IDE + GitHub Copilot) who want consistent rule behaviors across all assistants without manual copy-pasting. Open-Source Maintainers: Defining repo-wide architectural guidelines, testing rules, and security standards for community contributors using different AI tools. Engineering Teams & DevSecOps: Standardizing security guardrails, forbidden packages, and code styling rules across engineering teams, verifying rule consistency in CI/CD pipelines. ⚡ 1-Minute Quickstart Try RuleSync immediately in any project without installation: # 1. Bootstrap canonical AGENTS.md rule file and configuration npx rulesync init # 2. Lint your AI agent rules for conflicts, duplication, and token bloat npx rulesync lint # 3. Cross-compile target instruction files for all installed AI tools npx rulesync build 🔍 Why RuleSync? Feature Manual Copy-Paste Caliber / Paid Tools RuleSync (OSS) Cross-Tool Compilation ❌ Manual ⚠️ Limited ✅ 10+ Target Adapters Static Linter Engine ❌ None ❌ None ✅ 7 AST Linter Rules Conflict & Duplicate Detection ❌ None ❌ None ✅ AST Analysis Token Budget Optimization ❌ Manual ❌ None ✅ Automated Token Check Security & Secret Sanitizer ❌ None ❌ None ✅ Intercepts API Keys & Shell Bugs 100% Offline & $0 Operating Cost ✅ Yes ❌ Paid / SaaS ✅ Pure Local Open-Source CI / Pre-commit Synchronization ❌ None ❌ None ✅ rulesync check 🛠️ Supported Target Adapters Target Generated Output Description claude CLAUDE.md Anthropic Claude Code terminal agent rules cursor .cursor/rules/rulesync.mdc Cursor IDE rules with YAML frontmatter globs copilot .github/copilot-instructions.md GitHub Copilot repository guidelines cline .cline/instructions.json Cline / RooCode custom instructions JSON windsurf .windsurfrules Windsurf Cascade agent rules agents AGENTS.md OpenCode / Standard universal agent specification 🧪 Static Linter Rules Matrix RuleSync analyzes your markdown rules using AST parsing to catch common prompt anti-patterns before they cause LLM hallucinations: Rule Code Severity Description no-duplicate-rules error Flags duplicate or identical rule instructions. no-contradictory-directives error Catches conflicting guidance (e.g. prefer types vs prefer interfaces). no-unsafe-commands error Identifies dangerous shell commands (rm -rf /, chmod 777, curl | bash). no-secret-patterns error Intercepts hardcoded API keys (sk-ant-*, ghp_*, AKIA*) in rule files. max-token-budget warning Warns when rule sets exceed context window budgets (> 2,000 tokens). require-glob-scope warning Recommends scoping tech-specific rules with file glob annotations. no-empty-rules warning Flags empty rule header sections. 💻 CLI Reference rulesync init Bootstraps a canonical AGENTS.md file and creates .rulesync/config.json. npx rulesync init [--force] rulesync lint Lints source rule files against static analysis rules and token budgets. npx rulesync lint [--format json] [--max-tokens 2000] rulesync build Compiles source rules into all enabled target AI rule files. npx rulesync build [--targets claude,cursor,copilot] [--dry-run] rulesync check CI / pre-commit verification mode — exits with code 1 if target files are out of sync. npx rulesync check [--quiet] rulesync watch Live file watcher mode — automatically compiles target files whenever AGENTS.md is saved. npx rulesync watch rulesync diff Previews terminal colorized diffs between compiled target outputs and disk files. npx rulesync diff [--target cursor] ⚙️ Configuration (.rulesync/config.json) { "source": "AGENTS.md", "targets": ["claude", "cursor", "copilot", "cline", "windsurf"], "maxTokens": 2000, "headerComment": true, "ignoreRules": [] } 🔒 Security & Privacy Model Strict Workspace Boundary: Path resolution is strictly constrained within process.cwd() using sanitizeWorkspacePath(). Attempts to traverse outside workspace boundaries (../../) trigger an immediate security exception. 100% Offline & Private: Zero remote telemetry, zero external API calls, zero server tracking. 🤝 Contributing We welcome community contributions! Please read our Contributing Guide and Code of Conduct. 📜 License MIT License © 2026 RuleSync Contributors Topics Resources Readme MIT license Code of conduct Code of conduct Contributing Contributing Security policy Security policy Activity Stars 0 stars Watchers 0 watching Forks 0 forks Report repository