pro-workflow offers battle-tested Claude Code workflows designed to optimize the AI coding ratio, aiming for 80% AI-generated code and 20% human review. It targets developers seeking to enhance productivity through features like self-correcting memory, parallel worktrees, and structured wrap-up rituals for effective learning capture.

How It Works

The core approach centers on reducing AI code correction cycles via specific patterns. A self-correction loop enables Claude to learn from user edits automatically. Parallel worktrees ensure zero dead time by allowing users to work while Claude processes tasks. Wrap-up rituals formalize session endings, capturing key learnings. The system also emphasizes context discipline for managing token budgets and intelligent model selection for optimal accuracy.

Quick Start & Requirements

• Installation: Recommended via Claude Code plugin marketplace: claude plugin marketplace add rohitg00/pro-workflow followed by claude plugin install pro-workflow@pro-workflow. CLI installation is also supported.
• Prerequisites: Building TypeScript is required for persistent storage (npm install && npm run build in the plugin directory). Persistent learnings are stored in ~/.pro-workflow/data.db using SQLite.
• Links: Official documentation for agent teams is available at https://code.claude.com/docs/agent-teams.

Highlighted Details

• Version 1.1.0 introduces agent teams, custom subagents, adaptive thinking, smart commit with quality gates, and session insights.
• Smart commit functionality includes quality gates, code review, and learning capture.
• Session insights provide analytics on learning patterns, correction trends, and productivity metrics.
• Persistent storage leverages SQLite with FTS5 for full-text search of learned patterns and session data.
• Adaptive Thinking allows Opus 4.6 to calibrate reasoning depth per task.
• Hooks automate workflow pattern enforcement, such as quality gate reminders and post-edit checks.
• Supports multiple AI agent platforms via SkillKit integration.

Maintenance & Community

No specific details on maintainers, sponsorships, or community channels beyond the GitHub repository itself were found in the provided text.

Licensing & Compatibility

The provided README content does not specify a software license. This omission requires clarification for assessing commercial use or closed-source integration compatibility.

Limitations & Caveats

Agent Teams are noted as an experimental feature. The absence of explicit licensing information is a significant caveat for adoption and compatibility assessments.