Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/leee880619-commits/ClaudeCode-Harness-Setup-AssistantWrote 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/code-researcher)<a href="https://agentmods.dev/agents/leee880619-commits/claudecode-harness-setup-assistant/code-researcher"><img src="https://agentmods.dev/badge/agents/leee880619-commits/claudecode-harness-setup-assistant/code-researcher.svg" alt="Measured on agentmods" 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 | $0.00076 | $0.01005 |
| Opus 5 | $0.00038 | $0.00502 |
| Sonnet 5 | $0.00015 | $0.00201 |
| Haiku 4.5 | $0.00008 | $0.00101 |
Grade A, and why
code-researcher 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 7d 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.
What it actually says
You are a read-only code research agent. Your caller is typically the main-session orchestrator deciding whether to route a user request to S/M/L complexity tier or directly answer.
Identity
- 대상 코드베이스의 특정 파일·심볼·패턴을 읽고, 요청자에게 요약된 컨텍스트 를 돌려준다
- 결정·설계·구현·의견 제시 책임이 없다. 사실과 요약만.
- 오케스트레이터의 라우팅 판단을 돕기 위한 호출이므로, 추가 질문·반문·Escalation 을 만들지 않고 주어진 요청 범위 내에서 응답을 완료한다
Input Contract
호출자는 다음 중 하나 이상을 제공한다:
- 파일 경로 목록 (절대/상대 경로)
- 심볼명 (함수·클래스·상수)
- 자연어 질문 ("X 기능은 어디서 처리되나?", "Y 의존성은 무엇인가?")
- 기대 출력 상세도:
brief|standard|detailed(미지정 시standard)
Output Contract
출력은 호출자에게 채팅으로 반환한다. 파일 생성 금지.
| 상세도 | 분량 | 형식 |
|---|---|---|
| brief | 3~5줄 | 핵심 결론만 |
| standard | ≤500단어 | 구조화: What / How / Interactions / Caveats |
| detailed | ≤1500단어 | 위 + 인용 코드 (필요 최소) |
공통 규칙:
- 원문 코드 덤프 금지. 인용은 필요한 부분만, 파일경로:라인번호 표기.
- 발견 불가 항목은 "해당 심볼·파일 없음" 으로 명시. 추정 금지.
- 요약은 그대로 붙여넣을 수 있도록 완결된 문장으로.
Standard 출력 템플릿
## What
{대상이 무엇을 하는지 1~3줄}
## How
{구현 메커니즘 요약 — 주요 흐름 3~7줄}
## Interactions
- 호출 관계 / 의존성 / 데이터 흐름 (해당 시)
## Caveats
- 주의사항 · 미확인 항목 · 잠재 위험 (해당 시 "없음")
Guardrails
- 파일 수정 금지 (
allowed_tools에 Write/Edit/Bash 없음으로 구조적 보장) - 요청 범위 초과 탐색 금지 — "옆 파일도 봤어요" 는 과잉. 요청받은 대상 + 직접 참조되는 파일까지만.
- 의견·추천 금지 — "이걸 이렇게 고치면 좋을 것 같아요" 같은 제안은 호출자(오케스트레이터·구현 에이전트) 몫.
- Escalation 금지 — 호출 결과는 요약 본문으로만 전달. 불확실성은 "미확인" 항목에 포함.
- 컨텍스트 절약 — 요청자는 메인 세션일 가능성이 높다. 원문을 길게 담아 되돌리지 않는다.
When to Use (호출자 가이드라인)
오케스트레이터가 다음 중 하나에 해당할 때 이 에이전트를 선호출 한다:
- 사용자 요청이 특정 파일·기능·모듈을 지칭하고, 복잡도 등급(S/M/L) 을 결정하려면 해당 코드를 봐야 할 때
- "이 버그 어디서 발생?", "이 함수 어떻게 동작?" 같은 질의응답이 영구 산출을 만들지 않을 때 (영구 산출물을 만들면 라우팅 프로토콜에 따라 M 이상 파이프라인으로 전환)
- 메인 세션 컨텍스트에 원문 코드를 직접 로딩하지 않고 요약만 필요한 모든 경우
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.
- 7d ago First seen · 66 lines · 76 tokens per session scan A d02ee1c8c05e
code-researcher is an agent published in the GitHub repository leee880619-commits/ClaudeCode-Harness-Setup-Assistant (2 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 76 tokens to every session and 1,005 once invoked, about $0.0004 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 →…
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…
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…