Browse reusable Agent Skills, each with a clear purpose and practical guidance.
Shared return contract for SGAI Task subagents. Use before any delegated implementation, deployment, writing, verification, review, or project-critic subagent reports back to its caller.
Develop SubZeroClaw — the single-file ~550-line C agentic runtime. Load this before changing src/subzeroclaw.c or src/test.c: it carries the anti-framework thesis (the goal is NOT to grow), the code map (the loop, config, the shell tool, async compaction), what will and won't be merged, and the build/test loop. To merely run or configure SubZeroClaw, use subzeroclaw-use.
Run and operate SubZeroClaw — the ~550-line C agentic daemon (skill.md + LLM + shell + loop). Load this to build it, configure it (~/.subzeroclaw/config), write an agent skill, run a task, wire it to an unhardcoded router for routing/cache/compaction, or run it as a systemd service with credential isolation. WARNING: it executes arbitrary shell with no sandbox. To develop the runtime itself, use subzeroclaw-contribute.
Use when understanding your capabilities, how to use skills, or the rules of the Co-Researcher system.
Drive sgai (Software Garden AI) from any MCP-capable harness or AI agent. Covers the cyclical probe/poll/act workflow for managing AI software factory workspaces, sessions, and human interaction. Use this as the entrypoint when orchestrating sgai from Claude Code, Codex, or any AI harness.
Use when the user asks to "check white bear", "audit prohibitions", "find negative framing", or invokes /white-bear. Read-only audit: unnecessary competing-target mentions in LLM-facing prose.
Ralph Loop 병렬 버전. 여러 에이전트가 worktree에서 동시에 PRD 항목을 처리한다. 트리거: "병렬 실험", "autodev parallel", "동시에 실험", "워크트리 실험", "병렬 랄프" 안티-트리거: "순차 실험", "하나씩"
Context window management and token optimization guide. Triggers on: context, tokens, compact, memory management. NOT for: code writing, debugging.
에이전트 하네스 12원칙 기반 자가 진단 및 개선 제안. Triggers on: 하네스 진단, audit, 환경 점검, setup, maintenance, drift 확인. NOT for: 코드 구현, 버그 수정, 테스트 작성.