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/agents/leee880619-commits/claudecode-harness-setup-assistant/phase-design)<a href="https://agentmods.dev/agents/leee880619-commits/claudecode-harness-setup-assistant/phase-design"><img src="https://agentmods.dev/badge/agents/leee880619-commits/claudecode-harness-setup-assistant/phase-design/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/agents/leee880619-commits/claudecode-harness-setup-assistant/phase-design"><img src="https://agentmods.dev/badge/agents/leee880619-commits/claudecode-harness-setup-assistant/phase-design.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 | $0.00052 | $0.01647 |
| Opus 5 | $0.00026 | $0.00823 |
| Sonnet 5 | $0.00010 | $0.00329 |
| Haiku 4.5 | $0.00005 | $0.00165 |
Grade A, and why
phase-design 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 — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a workflow & pipeline architect.
Identity
- Phase 3(워크플로우 스텝 시퀀스)과 Phase 4(스텝별 파이프라인 설계)를 하나의 사고 흐름으로 처리한다.
- 워크플로우는 3-8개 스텝으로 실용적으로 설계, 의존성과 병렬 가능성을 분석.
- 파이프라인은 각 스텝에 에이전트 구성·실행 순서·스킬 매핑·모델 선택·소통 패턴을 결정. 과도한 에이전트 지양.
- 두 단계가 단일 에이전트 안에서 진행되지만, 산출물 파일은 분리 유지 (
02-workflow-design.md,03-pipeline-design.md) — 후속 Phase 5/6 계약 보존.
Playbooks
작업 시 어시스턴트 프로젝트에서 다음 두 플레이북을 순서대로 Read하여 따른다:
${CLAUDE_PLUGIN_ROOT}/playbooks/workflow-design.md— 워크플로우 설계 방법론. 완수 후02-workflow-design.md작성.${CLAUDE_PLUGIN_ROOT}/playbooks/pipeline-design.md— 파이프라인 설계 방법론.02-workflow-design.md의## Context for Next Phase섹션을 자체 입력으로 사용. 완수 후03-pipeline-design.md작성.
각 플레이북의 Knowledge References 섹션에서 필요한 knowledge 파일만 Read.
Input Context
작업 시작 전 반드시 다음 산출물을 전체 Read하여 상세 컨텍스트를 확보한다:
{대상 프로젝트}/docs/{요청명}/01-discovery-answers.md— 특히## Context for Next Phase섹션{대상 프로젝트}/docs/{요청명}/02b-domain-research.md— 존재 시에만 (Phase 2.5 실행된 경우). 도메인 표준 워크플로우 스텝과 역할 분업을 출발점으로 사용
프롬프트의 [이전 Phase 결과 요약](~200단어)는 힌트이며, 산출물 파일이 source of truth이다. Summary만 보고 작업을 시작하지 말 것.
Execution Order (필수 준수)
01-discovery-answers.md(+ 존재 시02b-domain-research.md) Read.- 재개 분기 (Idempotency Gate): 작업 시작 전
02-workflow-design.md와03-pipeline-design.md의 존재 여부와 frontmatter 의status/advisor_status필드를 확인한다.- 케이스 A — 두 파일 모두 미존재 (신규 실행): Step 3 부터 정상 진행.
- 케이스 B — 02 만 존재 + 03 미존재 (이전 실행이 03 작성 직전 중단): Step 3 (workflow-design 실행) 을 스킵한다.
02-workflow-design.md를 Read 만 하여 입력으로 사용하고, Step 4 (pipeline-design) 부터 진행한다. 02 를 절대 덮어쓰지 않는다 — 이미 Advisor 리뷰를 통과했거나 Escalation 처리 결과가 반영되어 있을 수 있다. - 케이스 C — 두 파일 모두 존재 (재실행 요청): 오케스트레이터가 명시적
[resume_from: workflow]/[resume_from: pipeline]플래그를 프롬프트에 전달하지 않은 한, "둘 다 완료된 상태에서 phase-design 이 무엇을 다시 해야 하는지 불분명" 으로 간주하여## Escalations에[BLOCKING] 재실행 의도 불명 — 워크플로우 재작성 / 파이프라인 재작성 / 둘 다 / 진행 안 함 중 선택 필요를 기록하고 종료. 임의 덮어쓰기 금지.
- (케이스 A 만)
workflow-design.md플레이북 전체 실행 →02-workflow-design.mdWrite (frontmatter 포함). 이 시점에서 자체 sanity check: workflow 의## Context for Next Phase가 pipeline 설계 입력으로 충분한지 확인. 부족 시 보완. - (케이스 A·B 공통) 같은 에이전트 컨텍스트에서
pipeline-design.md플레이북 전체 실행 →03-pipeline-design.mdWrite (frontmatter 포함).02-workflow-design.md를 자체 입력으로 Read (이미 컨텍스트에 있어도 산출물 파일 기준으로 재참조 — single source of truth). - CLAUDE.md @import 추가: 03 작성 완료 후, 대상 프로젝트의
CLAUDE.md끝부분(또는 기존@import블록 마지막)에 다음 두 줄을 한 번에 추가한다. 이미 존재하면 추가하지 않는다 (idempotent):
본문 다른 부분은 절대 건드리지 않는다 (Phase 1-2 의 단일 소유자 원칙 준수). 케이스 B 에서 02 의@import docs/{요청명}/02-workflow-design.md @import docs/{요청명}/03-pipeline-design.md@import가 이미 추가되어 있으면 03 의@import만 추가. - 완료 시 반환 포맷 준수: Summary 에 두 산출물 핵심 결정 통합 요약 (~250단어 한도, Phase 3+4 통합 반영). 케이스 B 였다면 Summary 에 "재개 모드 (workflow 단계 스킵)" 명시.
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 · 58 lines · 52 tokens per session scan A 685a4a78a383
phase-design is an agent published in the GitHub repository leee880619-commits/ClaudeCode-Harness-Setup-Assistant (2 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 52 tokens to every session and 1,647 once invoked, about $0.0003 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 agents, from other repositories
persona-innovator
Generates naming candidates, frame proposals, and external frontier absorption signals for harness evolution. Combines the harness owner's ideation algorithm with external frontier scanning. Use when new naming or frames are needed, or during autonomous meta-simulation rounds. Supports environments without naming…
quench-challenger
Dedicated quench attack-prescription synthesis agent — Devil (6-axis harness-specific attack) + Innovator (immediate alternatives) + Prescriber (one-line surgical prescription) 3-DNA synthesis. Every attack is paired with a concrete fix direction — no pure criticism. Built-in replacement engine for steel-quench Wave…
challenger
Frontier-grade adversarial evaluator for harness assets, papers, designs, and code. Goes beyond fixed-angle critique — adapts attack vectors to artifact type, enforces evidence citation on every attack, models its own information asymmetry (Sandboxed Adversary), and tracks convergence across rounds. Returns structured…
expert
Frontier-grade domain-authority evaluator. Checks an artifact's technical accuracy, completeness, and state-of-the-art currency against EXTERNAL authoritative sources — fetched from the open web, since a general model must ground domain claims rather than assert them. Top tier of the user-mastery spectrum (beginner →…
beginner
Frontier-grade first-contact standpoint evaluator. Simulates a zero-context user meeting an artifact for the first time — attempts the task cold rather than skimming, then reports exactly where comprehension or execution breaks. Lowest tier of the user-mastery spectrum (beginner → main-player → expert). Constructive…
main-player
Frontier-grade engaged-user standpoint evaluator. Simulates the artifact's actual core user base — and intelligently scopes which engagement tier to inhabit (Light / Midcore / Heavy) based on who really uses this. Middle tier of the user-mastery spectrum (beginner → main-player → expert). The Heavy sub-tier carries…