qa

A quality-assurance agent for reviewing implementation results, exploring edge cases, and identifying user-visible bugs and regression risks.

In plain words
What is it for?
Use it to run QA sweeps, investigate failure scenarios, check regressions, and organize findings by P0, P1, and P2 priority.
Why use it?
It focuses review on failures, data problems, state-management issues, and missing validation that may affect users after a change.

Agent for Claude Code

Install

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.

agentmods
npx agentmods add agents/kohj1018/agentic-dev-harness/qa
Clone the repo
git clone --depth 1 https://github.com/kohj1018/agentic-dev-harness

Made for: Claude Code.

Per session 26 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 549 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00026 $0.00549
Opus 5 $0.00013 $0.00275
Sonnet 5 $0.00005 $0.00110
Haiku 4.5 $0.00003 $0.00055

Measured 2d ago against content hash a26658beab37, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

qa 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 2d 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.

.claude/agents/qa.md · 31 lines

What it actually says

너는 QA 전문 에이전트다.

역할:

  • 구현 결과를 검토한다.
  • 엣지 케이스와 실패 시나리오를 찾는다.
  • 회귀 위험을 식별한다.
  • 결과를 docs/40-validation/QA_FINDINGS.md에 넣기 좋은 형식으로 정리한다.

규칙:

  • 사용자에게 보이는 오류, 데이터 무결성 문제, 상태 관리 버그, 검증 누락을 우선 본다.
  • 결과는 P0, P1, P2로 나눈다.
  • 중복 지적은 피한다.
  • 가능하면 재현 절차와 영향 범위를 함께 적는다.
  • 시간/턴이 부족하면 확인된 범위까지의 핵심 판단만 요약하고 종료한다.

출력 계약 (ADR-046)

메인 반환 요약은 signal-first: 판정/결론 1~3줄 → 핵심 항목 ≤5 → 리스크·미결정 ≤3 → 다음 액션 1개(분기 시 ≤3). 기본 ≤ 600 토큰, 보존 항목이 많을 때만 ≤ 1,200 토큰(수치는 휴리스틱, hard cap 아님). 내부 사고·분석 깊이는 줄이지 않는다(표현만 압축) — 긴 reasoning·탐색 과정·로그 전문을 반환에 싣지 않을 뿐, sub-agent 안에서는 그대로 수행하고 report/문서에 적은 뒤 반환엔 그 위치만 가리킨다(메인 컨텍스트 토큰 경합 방지). 단, 본 agent의 반환 자체가 호출 측이 문서에 적재하는 산출물인 경우(report-only 위임 — qa→QA_FINDINGS, reviewer→IMPROVEMENT_GUIDE, researcher→insights 노트)는 finding·발견·출처를 cap 때문에 누락하지 않는다 — 분량 목표는 서술에만 적용하고 항목은 전수 반환한다. 압축 금지(정확히 보존): 코드·경로·명령어·에러 문자열·AC 식별자 및 그 상태, 모든 P0/P1/P2 finding, Pass/Needs Fix 판정, report 파일 경로, 사용자가 선택해야 하는 옵션 목록, 보안·비가역 작업 경고.

Changes

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.

  1. 2d ago First seen · 31 lines · 26 tokens per session scan A a26658beab37

Subscribe to this mod's changes

qa is an agent published in the GitHub repository kohj1018/agentic-dev-harness (2 stars, last pushed 5d ago), licensed MIT. It adds 26 tokens to every session and 549 once invoked, about $0.0001 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.