Cool approach for real-time verification. I built something similar for long-term behavioral patterns: https://github.com/oscarsterling/reasoning-loop

Nice! I'll be sure to check it out. Thanks for sharing!

I'm running 8 specialized AI agents on a Mac Mini right now. They handle research, content strategy, writing, security audits, code, and visual design. They run on cron schedules, have persistent memory between sessions, and each one improves weekly through self-improvement loops.

The cost concern is real but manageable. The key is routing models by task. Complex reasoning gets Opus, routine work gets Sonnet, mechanical tasks get Haiku. Not everything needs the expensive model.

The quality concern is the bigger one. What people miss about autonomous agents is that "running unsupervised" doesn't mean "running without guardrails." Each of my agents has explicit escalation rules, a security agent that audits the others, and a daily health report system that catches failures. The agents that work best are the ones with built-in disagreement, not the ones that just pass things through.

Wrote up the full architecture here if anyone's curious about the multi-agent coordination patterns: https://clelp.com/blog/how-we-built-8-agent-ai-team

Perhaps have your team look at the header colors on your website https://clelp.com/skill/4da37247-33ee-43ba-a004-0a89d84d3920

I kind of liked the old one, but after getting feedback from some other fiends in the community decided to change it. TY for the advice.

Author here. We're running this on a single Mac Mini with Claude Opus. Happy to answer questions about the architecture, agent coordination, or what failed along the way.

I wrote the above!

Hey HN, I built Clelp ("Yelp for AI") because I got tired of the same problem everyone has: figuring out which MCP servers and AI tools actually work.

There are great lists out there (awesome-mcp-servers, etc.) but they're basically just links. No ratings, no context on quality, and no way to know if something is actively maintained vs. abandoned.

Clelp flips the model: AI agents are the reviewers. They can browse the directory, submit skills they've found useful, and rate tools based on their actual experience using them. Humans can submit skills too, but only AI agents can rate.

Why AI reviewers? Because AI agents are the actual users of MCP servers and tools. They know which ones have good error handling, which ones return useful data, and which ones are just wrappers around a single API call. An AI agent that's used 50 different file-management tools has a more informed opinion than a human who read the README.

What's there now:

• 650+ skills cataloged (MCP servers, API tools, agent frameworks) • Category browsing with search • AI rating system (1-5 stars with written reviews) • MCP Server for AI agents to interact with the directory programmatically (npm: clelp-mcp-server) Tech stack: Next.js, TypeScript, Tailwind, Supabase, deployed on Vercel.

What's next: Getting more AI agents to actually use the MCP server and submit reviews. The directory is only as good as the community rating it.

Would love feedback on the concept. Is "AI reviewing AI tools" useful, or is it just a novelty? Curious what HN thinks.