Pentest AI Agent — 발견과 부재를 모두 증명합니다. Red-teaming LLM Agent: chatbots, RAG Assistants, Tool-calling copilots, MCP 서버 및 자율 ReAct Agents · 24 기술 (23 사냥 + 방법론) · 4 탐지 팔 · 24 / 24 실험실 게이트 · 7 대상 환경 · FN 0 오버 949 필드 실행 · Claude Code, OpenCode, Hermes, Codex CLI 및 AntiGravity로로드하십시오.

Sachin Sharma - 버그 Hunting & GenAI 보안 연구.

---

Claude-BugHunter에 Sibling — 동일한 모양, 다른 채석장. BugHunter는 웹 앱을 사냥합니다. Claude-AgentHunter는 에이전트를 스스로 사냥합니다.

이것은 무엇인가

Claude Code 기술 시스템에 대한 드롭 인 기술 번들 (및 OpenCode, Hermes, Codex CLI, AntiGravity). 한 번 설치 하네스는 에이전트 보안을 추측 하 고 훈련 된 pentest 실행 시작: 그것은 대상 에이전트를 지문, 오른쪽을 선택 hunt-agent-*] 표면에 의해 기술, 그들의 stdlib-only 프로브 실행, 4 팔 엔진과 추적을 점수, 그리고 게이트를 모든 발견 그래서 유창한 답변을 결코 누출으로 발송 하지 않고 침묵 프로브는 결코 “깨끗”.

Claude-AgentHunter

Pentest AI agents — and prove both the findings and the absences. A skill bundle plus a runnable engine for red-teaming LLM agents: chatbots, RAG assistants, tool-calling copilots, MCP servers, and autonomous ReAct agents · 24 skills (23 hunt + methodology) · 4 detection arms · 24/24 lab gates · 7 target environments · FN 0 over 949 field runs · loads into Claude Code, OpenCode, Hermes, Codex CLI and AntiGravity.

Built by Sachin Sharma — Bug Hunting & GenAI Security Research.

---

Sibling to Claude-BugHunter — same shape, different quarry. BugHunter hunts web apps; Claude-AgentHunter hunts the agents themselves.

24 skills   ·   4 detection arms   ·   24/24 lab gates   ·   7 target environments   ·   FN 0 over 949 runs

What is this

A drop-in skill bundle for the Claude Code skills system (and OpenCode, Hermes, Codex CLI, AntiGravity). Install once and the harness stops guessing at agent security and starts running a disciplined pentest: it fingerprints the target agent, picks the right hunt-agent-* skills by surface, runs their stdlib-only probes, scores the traces with a 4-arm engine, and gates every finding so a fluent answer never ships as a leak and a silent probe never ships as “clean”. Describe the target in plain English and the relevant skills auto-load — no invocation by name. The whole flow runs as one command, /hunt-agent, or stage by stage (see Commands).

Why it exists

An agent generates text; it does not retrieve state. That creates two failure modes a web pentest never has to worry about, and Claude-AgentHunter is built around catching both:

• A fluent answer looks identical to a real leak → the False-Positive gate.
• A silent probe looks identical to a secure target → the Negative-Result gate.

The second one is the hard one. Across this project’s own development, ten probe defects each reported a compromise as clean — a similarity gate that scored a real leak 0.03, an HTTP 500 counted as a refusal that buried a confirmed bypass, a token floor that hid the forgery that leaked the flag. Every one looked like a passing test. Both gates live in engine/gates.py and ship in every skill.

How the skills work

Each skill is skills//SKILL.md — Agent-Skills markdown with frontmatter (name, description, sources, validated_against, report_count — the count of in-house field-validated reports behind the skill, 0 for reference-only). Harnesses topic-match on the description; describing your test in plain English loads the relevant skills, no explicit invocation. Unlike static methodology skills, each one carries a Runnable Probe that executes against a live agent and a