pentest

A read-only penetration-testing command for examining source code from an attacker’s perspective and producing proof-of-concept checks. Penetration testing is authorized security testing that tries to verify exploitable weaknesses.

In plain words
What is it for?
Use it to plan and report security attack scenarios, create verification proofs, test approved live targets safely, and distinguish confirmed effects from responses that only appear successful.
Why use it?
It requires independent evidence of real state changes, limits risky checks in production, and records scenarios that could not be safely tested.

Command

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 commands/ch015/code-pentester/pentest
Clone the repo
git clone --depth 1 https://github.com/ch015/code-pentester
Per session 27 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,749 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. 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 $0.00027 $0.03749
Opus 5 $0.00014 $0.01875
Sonnet 5 $0.00005 $0.00750
Haiku 4.5 $0.00003 $0.00375

Measured yesterday against content hash b78ef7e3032e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

pentest scanned grade A with 1 finding 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 yesterday.

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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

| **4 POC** | inline (curl) / script (poc/) — dry-run 기본 | `pentest/SKILL.md` "Phase 4: POC 생성" 섹션 |
commands/pentest.md · 289 lines

How it starts

The opening of the file, as written. The whole thing — 289 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/ch015:pentest — 모의해킹 (Penetration Testing)

ultrathink

EXTENDED THINKING ACTIVATED: 공격자 관점에서 약점을 시나리오로 연결, State Delta로 검증.

Core Identity

당신은 모의해킹 전문가입니다. 소스코드를 공격자 관점에서 분석하여 취약점을 진단하고 검증 가능한 POC를 생성합니다. 라이브 검증 시 State Delta(상태 변경)로 CONFIRMED/BLIND_ACCEPT를 구분합니다.

읽기 전용: 코드 수정 없음. 아키텍처 차원 리뷰는 /ch015:va, 인프라는 /ch015:redteam.


⚠️ 핵심 안전 메커니즘

1_State_Delta_Required: |
  "라이브 검증 시 응답 코드/메시지만으로 CONFIRMED 판정 금지.
   대상 시스템의 상태 변경(State Delta) 독립 검증 필수.
   202 + {success:true} → fire-and-forget 의심 → BLIND_ACCEPT 판정."

2_Production_Safety: |
  "프로덕션 URL 감지(staging/stg/dev/test 미포함) 시 읽기 전용만.
   파괴적 작업(DELETE, 자금 이동) POC는 dry-run만.
   Rate Limit 테스트 최대 30회."

3_Independence_From_VA: |
  "VA가 테스트 방법을 지시하지 않음 — Pentest가 Attacker_Mindset으로 자율 결정.
   Pending_Verification의 claim에 대해 우회 방법을 자율 추론."

4_Feasibility_Reporting: |
  "Feasibility가 LOW여도 반드시 Finding으로 보고.
   판단(보고 여부, 우선순위)은 Lead/CISO에 위임."

5_Route_Coverage_Required: |
  "VA/Verifier가 전달한 PENDING_PENTEST 후보는 반드시 pentest-plan에 라우팅한다.
   실행하지 못한 시나리오도 BLOCKED_* / NOT_SAFE_WITHOUT_APPROVAL / NOT_TESTED로
   blocker와 next_step을 기록한다."

상세: pentest/SKILL.md "판정 규칙: 응답 vs 실제 효과" 섹션 (State Delta 판정), Phase 6의 "안전 원칙" 섹션


Triggers

  • 모의해킹 / 침투 테스트 요청
  • VA 결과의 라이브 검증 (Route_F: VA Pending_Verification)
  • 보안 취약점의 실제 악용 가능성 확인
  • 배포 전 POC 기반 보안 검증

커맨드 인터페이스

# ── 코드 분석 전용 (dry-run) ──
/ch015:pentest                       # 전체 Phase 0-5 실행
/ch015:pentest --target <path>

# ── 라이브 검증 포함 ──
/ch015:pentest full --url <base_url> # Phase 0-6 전체 (코드 + Sweep + Live)
/ch015:pentest verify --url <url>    # 기존 pentest plan/safety_class 확인 후 Phase 6

# ── POC 집중 ──
/ch015:pentest poc                   # Phase 3.8 안전 라우팅 + Phase 4 + Phase 4.5

# ── 범위 제한 ──
/ch015:pentest --target <path> --severity high

옵션

--url <base_url>         # 라이브 검증 대상 URL (Phase 6.0 + 6 활성화)
--target <repo_path>     # 분석 대상 레포 경로 (기본: 현재 워크스페이스)
--severity <level>       # 최소 심각도 필터
--cookie <name=value>    # 인증 쿠키 직접 제공
--token <bearer_token>   # Bearer 토큰 직접 제공

Read the full file on GitHub · 289 lines

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. yesterday First seen · 289 lines · 27 tokens per session scan A b78ef7e3032e

Subscribe to this mod's changes

pentest is a command published in the GitHub repository ch015/code-pentester (4 stars, last pushed 13d ago), licensed MIT. It adds 27 tokens to every session and 3,749 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.