claude-code-workflows and claude-code-config
These are complements: workflows provide the execution framework and automation patterns, while governance policies provide the contextual rules and constraints that determine *how* those workflows should behave in specific situations.
About claude-code-workflows
shinpr/claude-code-workflows
Production-ready development workflows for Claude Code, powered by specialized AI agents.
Provides modular, multi-agent orchestration for Claude Code with specialized agents handling requirements analysis, design documentation, implementation, and quality verification across backend, frontend, and fullstack workflows. Includes optional plugins for product discovery, behavior monitoring, and deployment governance, plus a skills-only mode for integration into existing orchestration systems. Supports diagnostic and reverse-engineering workflows to investigate bugs and document legacy codebases.
About claude-code-config
aaronsb/claude-code-config
Event-driven policy, process, and governance for Claude Code. Ways encode how we do things — prescriptive rules triggered by context, not requested by intent.
Implements just-in-time policy injection via hook-based semantic matching (embedding or BM25) that fires before tool execution, maintaining adherence as context grows by placing guidance near the attention cursor rather than in the system prompt. Hooks intercept user messages, commands, and subagent spawning to trigger domain-specific ways once per session, with tiered matching combining regex patterns for known keywords/files alongside semantic scoring. Designed for Claude Code but framework-agnostic—extensible to any workflow domain via `ways.json` configuration and custom `way.md` files in `~/.claude/hooks/`.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work