Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add leee880619-commits/ClaudeCode-Harness-Setup-Assistant/plugin install harness-architectWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/leee880619-commits/claudecode-harness-setup-assistant/harness-setup)<a href="https://agentmods.dev/commands/leee880619-commits/claudecode-harness-setup-assistant/harness-setup"><img src="https://agentmods.dev/badge/commands/leee880619-commits/claudecode-harness-setup-assistant/harness-setup/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/leee880619-commits/claudecode-harness-setup-assistant/harness-setup"><img src="https://agentmods.dev/badge/commands/leee880619-commits/claudecode-harness-setup-assistant/harness-setup.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00035 | $0.08906 |
| Opus 5 | $0.00017 | $0.04453 |
| Sonnet 5 | $0.00007 | $0.01781 |
| Haiku 4.5 | $0.00003 | $0.00891 |
Grade A, and why
harness-setup scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 341 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Harness Architect — Orchestrator
대상 프로젝트를 분석하여 Claude Code 하네스(CLAUDE.md, settings, rules, agents, playbooks, hooks, MCP)를 9단계로 구축합니다. 이 커맨드는 메인 세션(Orchestrator) 역할만 수행하며, 실질 작업은 서브에이전트(subagent_type)에 위임합니다.
세션 시작 규칙 로딩
아래 4개 규칙 파일은 .claude/rules/ 아래에 YAML frontmatter 없이 배치되어 있어 Claude Code 플러그인 런타임이 always-apply로 자동 로딩합니다. 별도 Read 호출이 필요하지 않습니다 — 오케스트레이터 세션 시작 시점에 이미 컨텍스트에 들어있다고 가정하고 진행하세요.
참고 파일 목록 (진단·수정 목적으로 참조 시 경로):
${CLAUDE_PLUGIN_ROOT}/.claude/rules/orchestrator-protocol.md— Phase 전환/에스컬레이션/피드백 프로토콜${CLAUDE_PLUGIN_ROOT}/.claude/rules/question-discipline.md— AskUserQuestion 사용 규율${CLAUDE_PLUGIN_ROOT}/.claude/rules/output-quality.md— 생성물 품질/보안 기준${CLAUDE_PLUGIN_ROOT}/.claude/rules/meta-leakage-guard.md— 메타 누수 방지 가이드
만약 always-apply 로딩이 작동하지 않는 환경(예: 플러그인 로더 미지원 버전)이면 해당 파일을 직접 Read하되, 일반적으로는 중복 로딩을 피합니다.
핵심 개념 정의
| 개념 | 정의 |
|---|---|
| 워크플로우 | 작업 단계 시퀀스 (Research → Design → Implement → QA → Deploy) |
| 파이프라인 | 각 스텝의 에이전트 실행 체인 (병렬/순차) |
| 에이전트 팀 | TeamCreate/Agent/SendMessage 기반 소환·소통 그룹 |
| 에이전트 (WHO) | 페르소나·목적·규칙 정의 (lean ~25줄) |
| 스킬/플레이북 (HOW) | 방법론·도구·템플릿 정의 (detailed) |
Agent-Playbook 분리 원칙
- WHO:
${CLAUDE_PLUGIN_ROOT}/.claude/agents/*.md— 에이전트 정체성 - HOW:
${CLAUDE_PLUGIN_ROOT}/playbooks/*.md— 방법론. 메인 세션의 Skill 도구로 노출되지 않도록playbooks/에 둔다. 서브에이전트가 Read하여 실행.
핵심 원칙
- AskUserQuestion은 Orchestrator 전용 — 서브에이전트는 Escalations에 기록, Orchestrator가 취합.
- 암묵적 합의 금지 — 모든 설정은 명시적 답변 또는 스캔 결과로 결정.
- Target Project Guardrail — 대상 프로젝트 작업 중에는 플러그인/본 커맨드 파일을 수정하지 않음.
- No Meta-Leakage — 생성 파일에 본 플러그인의 행동 규칙을 포함하지 않음.
- Playbook 직접 실행 금지 — 방법론은 반드시 Agent 도구로 서브에이전트(opus)를 소환해 실행.
금지: Skill(skill: "fresh-setup")
필수: Agent(subagent_type: "harness-architect:phase-setup", description: "...", prompt: "...")
Orchestrator Architecture
메인 세션(Orchestrator):
- Phase 0: 경로 수집 + 기존 작업 감지 + 요청명 생성
- 라우팅: 대상 프로젝트 상태에 따라 첫 에이전트 결정
- Phase 전환: 에이전트 결과 수신 → Escalations 처리 → 다음 에이전트 소환
- 진행 피드백: "Phase N/9: {이름}" 표시
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 8d ago First seen · 341 lines · 35 tokens per session scan A 6bd0e711a102
harness-setup is a command published in the GitHub repository leee880619-commits/ClaudeCode-Harness-Setup-Assistant (2 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 35 tokens to every session and 8,906 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
send
Send a message to a running agent session. Use this to correct or direct a live agent mid-stream without killing and respawning it.
discover
Search the Everruns(Dev) API catalog for available operations.
me-figure-discussion
Write or review thermal-fluid figure discussion with observation, physical mechanism, comparison, uncertainty, and claim limits.
pick-agent
A command that helps users choose and import only the agent role they need, such as a frontend developer, researcher, or editor.
agent-wizard
Build a Claude Agent SDK agent by answering 9 illustrated questions, then run it.
cleanup
Run a systematic cleanup on the provided code to resolve smells, simplify complexity, and enforce best practices.