Run a Karpathy-style planner/generator/evaluator agent loop in your repo: a manager negotiates a testable contract, then drives a resumable...
#claude-code
2275 posts
Run one request through several Claude Code skills in parallel isolated subagents, then merge or judge the best answer.
Evaluate your startup idea from 1 to 5
Auto-register for Luma (lu.ma) events — a Claude Code skill that fills forms from a saved profile, drafts grounded answers from your GitHub,...
A package manager for your agent skills — one canonical library, loaded on demand, never all at once.
HubWu 精选开源 Agent Skills 合集 —— 跨 agent 通用(遵循 agentskills.io 开放标准,Claude Code / Codex / Gemini CLI / Cursor 等),持续收录实...
Write your dev workflow once. Run it on any LLM. Open agent skills + multi-model swarm + a local findings store that learns which models you...
Deterministic, zero-dep routing linter for a skill catalog — catch description collisions and scope overclaim before you ship a skill. Ports...
Make AI coding agents write RAPIDS-first GPU Python.
Conformance suite for Agent Skills: lint SKILL.md against the spec, run against real agent runtimes, publish a compatibility matrix
Read-only Kalshi + Polymarket market data skill for AI agents (Claude Code, Codex, Cursor). Zero dependencies, no API keys, auditable in 5 m...
Public agent skills for Codex and Claude Code