
BowTiedBrogrammer
928 posts



This is the most complete Claude Code setup that exists right now. 27 agents. 64 skills. 33 commands. All open source. The Anthropic hackathon winner open-sourced his entire system, refined over 10 months of building real products. What's inside: → 27 agents (plan, review, fix builds, security audits) → 64 skills (TDD, token optimization, memory persistence) → 33 commands (/plan, /tdd, /security-scan, /refactor-clean) → AgentShield: 1,282 security tests, 98% coverage 60% documented cost reduction. Works on Claude Code, Cursor, OpenCode, Codex CLI. 100% open source.


there's still a ton of opportunity out there even with all this AI stuff because people always overestimate downside and underestimate upside think about ads. you spend $100 to test something. worst case you lose $100, best case you find an angle worth 10x more but most people judge experiments the wrong way 🙉: count the NUMBER of failures, worried about the win rate 🦊: count the payoff luckily that miscalculation is where a lot of opportunity comes from if everyone else is too scared to run cheap tests, the people who understand the numbers get the upside there's whole businesses around pricing risk better than other people: insurance, warranties, guarantees. a lot of money gets made by understanding risk better than the average person so I wonder why people avoid risk for experiments. maybe it's because school averages your past failures into your final grade but in real life, you can keep taking small shots until one hits. and one good hit can pay for a long list of misses








It's been years and I still refuse to ever use this emoji (OGs remember)












