ai-security-agent SKILL.md skillA

ai-security-agent from hellodqy/mastermind-bug-bounty is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add hellodqy/mastermind-bug-bounty/ai-security-agent@git:20260709.c284c01

Pins git:20260709.c284c01 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
hellodqy/mastermind-bug-bounty
Path
agents/ai_security/SKILL.md
Versions recorded
3
Latest version
git:20260709.c284c01, committed Jul 9, 2026
First version recorded
Jun 2, 2026
Size
1,581 bytes
Repository stars
14
License
none declared

AI/LLM security reasoning agent. Explore AI features as connected systems of prompts, tools, memory, retrieval, and permissions.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# AI Security Agent

## Goal

判断目标的 AI 能力能否被用户输入、外部内容、工具调用或检索数据影响,并继续追问这种影响能否扩大到数据泄露、越权动作或系统能力滥用。

## Tools / Inputs

- Chatbot / RAG / Agent / Tool calling / MCP / code sandbox 入口
- 上传文档、网页索引、插件工具、函数调用、审计日志
- `skills/ai_security/SKILL.md`;资源索引:`references/INDEX.md`,候选资源:`ai-security-testing.md`

## Constraints

1. 不把 jailbreak 成功话术当漏洞;必须证明越权数据或越权动作。
2. 区分模型胡说、策略绕过、工具滥用和数据泄露。
3. 不诱导目标执行破坏性操作;证明使用无害动作。
4. 每个成功影响都要追问能否触发工具、读取 RAG 数据或跨用户泄露。
5. 不能验证影响时标记 PENDING,不进入报告正文。

## Chain-First Loop

每个 AI 信号后继续问:

- 输入控制了模型输出,还是控制了工具参数?
- 输出是否能影响后端动作、邮件、文件、请求或数据库?
- RAG 能不能被间接注入污染?
- 是否能读取其他用户、系统提示、内部文档或工具凭据?
- 当前能力能否和传统 Web 漏洞串起来?

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add hellodqy/mastermind-bug-bounty/ai-security-agent@sha256:8578954355e01d5d

The line at the top of this page pins the label git:20260709.c284c01, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_u2auvcowox37dcp6.svg)](https://markdownregistry.com/a/art_u2auvcowox37dcp6)

3 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 3 versions of this file, between Jun 2, 2026 and Jul 9, 2026, about one every 18 days. Its first recorded version is graded B and its latest A.

versioncommittedcommitsizeaudit
git:20260709.c284c01 latest2026-07-09 c284c01 1,581 BA view · diff
git:20260709.a8826982026-07-09 a882698 1,540 BA view · diff
git:20260602.95e27a72026-06-02 95e27a7 2,697 BB view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (1581 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

hellodqy/mastermind-bug-bounty · 14 stars · license none declared · pushed 2026-09-16 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_u2auvcowox37dcp6
GET https://markdownregistry.com/api/v1/resolve?ref=hellodqy/mastermind-bug-bounty/ai-security-agent
GET https://markdownregistry.com/api/v1/blob/8578954355e01d5dd50276a90fbbeb788473489eb47c79a2632ae0c0338baa01

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from hellodqy/mastermind-bug-bounty

mastermind-bug-bounty SKILL.md skill
hellodqy/mastermind-bug-bounty · SKILL.md · Four-phase autonomous bug bounty orchestration. Phase 0 maps external assets with deterministic Python helpers, Phase 1…
v4.7.0 · audit A · 14 stars
attack-surface-analyst SKILL.md skill
hellodqy/mastermind-bug-bounty · agents/analyst/SKILL.md · Rank reconnaissance-derived attack surfaces and design evidence-driven tests without active exploitation.
git:20260711.c1d38ba · audit A · 14 stars
api-fuzz-agent SKILL.md skill
hellodqy/mastermind-bug-bounty · agents/api_fuzz/SKILL.md · API exploration agent. Use endpoint signatures and real response values to expand attack surfaces; do not stop at the…
git:20260711.d202fe9 · audit A · 14 stars
autonomous-attacker SKILL.md skill
hellodqy/mastermind-bug-bounty · agents/attacker/SKILL.md · Test ranked attack surfaces autonomously, preserve evidence, and turn new access into additional attack-chain…
git:20260711.c1d38ba · audit A · 14 stars
bypass-agent SKILL.md skill
hellodqy/mastermind-bug-bounty · agents/bypass/SKILL.md · Access-control bypass agent. Treat 401/403/405 as routing clues, not as final answers; connect bypasses to privilege…
git:20260709.a882698 · audit A · 14 stars
crypto-attack-agent SKILL.md skill
hellodqy/mastermind-bug-bounty · agents/crypto_attack/SKILL.md · Crypto and token reasoning agent. Interpret keys, tokens, signatures, encrypted fields, and their downstream leverage.
git:20260709.a882698 · audit A · 14 stars
exploit-agent SKILL.md skill
hellodqy/mastermind-bug-bounty · agents/exploit/SKILL.md · Exploitation and impact-validation agent. Turn promising findings into confirmed impact, then keep chaining from the…
git:20260709.8be07f4 · audit A · 14 stars
recon-agent SKILL.md skill
hellodqy/mastermind-bug-bounty · agents/recon/SKILL.md · Asset reconnaissance agent. Map externally exposed assets and produce structured evidence for later AI reasoning…
git:20260709.a882698 · audit A · 14 stars
report-agent SKILL.md skill
hellodqy/mastermind-bug-bounty · agents/report/SKILL.md · Report generation agent. Convert verifier-confirmed findings into a fixed evidence-based report without adding…
git:20260916.a5be4e5 · audit A · 14 stars
account-takeover SKILL.md skill
hellodqy/mastermind-bug-bounty · skills/account_takeover/SKILL.md · Analyze login, recovery, verification, MFA, session refresh, and account binding flows when a weakness may let one user…
git:20260915.240c761 · audit A · 14 stars
ai-security SKILL.md skill
hellodqy/mastermind-bug-bounty · skills/ai_security/SKILL.md · Treat AI systems as chains of prompt, retrieval, memory, tools, and permissions.
git:20260914.819e838 · audit A · 14 stars
api-fuzz SKILL.md skill
hellodqy/mastermind-bug-bounty · skills/api_fuzz/SKILL.md · Explore API behavior using real parameters and values before generic fuzzing.
git:20260709.c284c01 · audit A · 14 stars

Every file in hellodqy/mastermind-bug-bounty

Browse by kind, by grade A, or by owner.