Bee2Bee

Open-source software powers the digital economy, yet 96% of repository issues remain unresolved after 90 days. Contributors struggle to find issues matching their skills, while maintainers waste hours triaging and matching tasks to developers. We've built an autonomous multi-agent AI system that transforms this broken process: contributors chat naturally with specialized AI agents that analyze repositories using MeTTa knowledge graphs, intelligently match skills to issues, and automate instant bounty payments in PayPal USD (PYUSD). Instead of manually searching through hundreds of repos, developers simply ask "show me Python issues I can solve" and our agents—RepoAnalyzer, SkillMatcher, and BountyEstimator—collaborate to find perfect matches with transparent reasoning about why each issue fits their expertise.

Bee2Bee is built with Next.js 14 + TypeScript frontend using shadcn/ui components, connected to Supabase (PostgreSQL) for data persistence and auth. The core innovation is our multi-agent architecture using Fetch.ai's uAgents framework where autonomous agents analyze GitHub repositories, combined with SingularityNET's MeTTa symbolic reasoning engine to classify repository complexity, tech stacks, and difficulty levels through logical inference rather than ML models.