THIS IS THE MOST COMPLETE CLAUDE CODE SETUP I'VE EVER SEEN someone open-sourced a full agent harness for Claude Code 38 agents. 156 skills. 72 commands. all pre-configured the kind of setup you'd spend weeks building yourself what's inside: → planner - feature implementation planning → architect - system... design decisions → tdd-guide - test-driven development workflow → code-reviewer - quality and security review → security-reviewer - vulnerability analysis → build-error-resolver - fixes broken builds → refactor-cleaner - dead code cleanup → doc-updater - keeps docs in sync → chief-of-staff - communication triage → + 29 more agents 156 skills covering token optimization, memory persistence, continuous learning, verification loops, parallelization, subagent orchestration it ships with AgentShield baked in a security auditor that runs vulnerability checks on every agent action before execution works on Claude Code, Cursor, OpenCode, and Codex CLI one repo, one install, configured everywhere one warning: loading 38 agents and 156 skills into your context at once will burn your usage limits in minutes pick the 3-4 you actually need and load only those free. open-source. MIT licensed like + bookmark - you'll need thisshow more

Great Work! @affaan Here is a visual map + narrated walkthrough of everything-claude-code

I run 9 specialised agents on enterprise codebases. At that count, the challenge is not agent capability. It is decomposition. Each agent needs a distinct, non-overlapping scope with explicit file ownership before it starts. I had to lose real work to a concurrent write before that became a hard rule. The agent count is not the bottleneck. The task decomposition is.

bro really open sourced the whole thing thats insane

Looks interesting for sure. Thanks

thanks too

This is hardly impressive these days.

why?

Incredible setup! So many agents and skills

Your warning at the end is the most important line. Running 5 specialized agents daily — learned this the hard way. Started with 12 skills per agent, context windows hit 300K tokens and responses turned to mush. Cut each agent down to 3-4 core skills + one SOUL.md identity file. Performance jumped overnight. The real architecture insight: don't build one super-agent with 38 skills. Build 5 focused agents that each own one domain. My "fire" agent handles content, "earth" handles infra, "gold" handles strategy. They never step on each other's context.

and all of this is in the public domain…

The setup is impressive. The harder question is which of those skills still hold up once the workflow has to survive maintenance instead of a weekend demo.

LETSS GOOO CLAUDE COODE

Great..

and it’s fully open sourced? no way..

This is insane, saves weeks of tinkering. Huge props to the folks who put this together, can't wait to dig in and play with those agents.

already been running this for a week. the planner agent alone is wild