H-mmer/pentest-agents

---

name: hunt-business-logic

description: Hunting skill for business-logic vulnerabilities (CWE-840 Business Logic Errors, CWE-841 Improper Enforcement of Behavioral Workflow, CWE-639 Authorization Bypass via User-Controlled Key in business contexts, CWE-362 race conditions on financial flows). Built from 44 corpus reports plus 8.8K shared-platform reports across HackerOne, Bugcrowd, Huntr, GitHub Security Advisories, plus 2024-2026 meta verified against NVD — Lilishop coupon overpurchasing (CVE-2024-50654 CVSS 7.5), WWBN AVideo wallet double-spend TOCTOU (CVE-2026-34368, GHSA-h54m-c522-h6qr), Keycloak 2FA bypass (CVE-2025-3910, GHSA-5jfq-x6xp-7rw2), AlegroCart 1.2.9 negative-quantity price manipulation (Andrey Stoykov SecLists Apr 2025), Bagisto cart price manipulation (Rudransh Singh Rajpurohit Sep 2025), Doppler free-trial reset (Aditya Sunny Dec 2024), Stripe hasEverTrialed bypass (better-auth

---

name: hunt-idor

description: Hunting skill for Insecure Direct Object Reference / Broken Object Level Authorization (BOLA — OWASP API1:2023). Built from 1,117 public IDOR bug bounty reports across HackerOne, Bugcrowd, Intigriti, GitHub Security Advisories, Huntr, plus 2024-2026 meta verified against NVD — Sam Curry's automotive chain (Kia/Hyundai/Ferrari), CVE-2026-30956 (OneUptime tenant header bypass CVSS 9.9), CVE-2025-64431 (Zitadel V2Beta Org API), CVE-2026-32131 (Zitadel Management API), CVE-2025-61876 (Inforcer tenant enumeration), CVE-2024-45719 (Apache Answer UUIDv1 token prediction), CVE-2024-50633 (Indico BOLA), plus the GraphQL field-level / nested-object pivot wave and agentic AI cross-tenant family (FastGPT, WeKnora, Paperclip). Use when hunting IDOR / BOLA / cross-tenant data access on web apps, APIs, GraphQL endpoints, multi-tenant SaaS, mobile apps, automotive/IoT platforms, and AI/ML inference servers.

---

name: hunt-info-disclosure

---

name: hunting-methodology

---

name: hunt-llm-ai

description: Hunting skill for LLM and Agentic AI vulnerabilities — direct + indirect prompt injection, ASCII smuggling data exfil, agentic tool-use abuse, system prompt leakage, vector DB cross-tenant, model server RCE, insecure output handling. Built from public bug bounty reports across HackerOne, Huntr, Project Zero, GitHub Security Advisories, plus 2024-2026 meta verified against NVD — Microsoft 365 Copilot ASCII Smuggling (Johann Rehberger Aug 2024 disclosure), CVE-2025-46059 LangChain GmailToolkit indirect prompt injection (CVSS 9.8), CVE-2025-68613 LangChain PythonREPLTool semantic RCE (CVSS 9.8), CVE-2024-46946 LangChain LLMSymbolicMathChain sympy.sympify, CVE-2025-27520 + CVE-2025-32375 + CVE-2024-2912 BentoML pickle, Ollama RCE family (CVE-2024-37032, CVE-2024-39722, CVE-2024-45436, CVE-2025-44779), CVE-2025-64496 Open WebUI Direct Connections SSE code injection (GHSA-cm35-v4vp-5xvx), CVE-2024-1483/1560/1594 MLflow path traversal. Covers OWASP LLM Top 10 v2025 (LLM01-LLM10) and OWASP Agentic AI Top 10 (AA-01 through AA-10). Use when hunting prompt injection, jailbreaks, agentic tool abuse, RAG poisoning, vector DB IDOR, MCP server compromise, model registry RCE, or any "AI feature" surface in a target.

---

name: hunt-oauth

description: Hunting skill for OAuth 2.0 / 2.1, OpenID Connect (OIDC), SAML SSO, and JWT authentication. Built from 365 public OAuth/SSO bug bounty reports across HackerOne, GitHub Security Advisories, PortSwigger Research, GitHub Security Lab, Detectify, Doyensec, Salt Labs, Semperis, Obsidian Security, Trace37 plus 2024-2026 meta verified against NVD — ruby-saml parser differentials (CVE-2025-25291/25292, GHSL-2024-329/330), Authentik regex redirect_uri (CVE-2024-52289), workers-oauth-provider PKCE downgrade and redirect_uri (CVE-2025-4143/4144), Entra ID actor token cross-tenant impersonation (CVE-2025-55241), Hono JWT alg confusion (CVE-2026-22817), nOAuth omniauth-microsoft_graph (CVE-2024-21632), Tekton git resolver token exfil (CVE-2026-40161), Flux Operator OIDC empty claims (CVE-2026-23990), Argo CD project token (CVE-2025-55190), tinyauth OIDC client binding (CVE-2026-32245). Use when hunting OAuth / OIDC / SAML / JWT / SSO bugs on web apps, mobile apps, single-page apps, MCP servers, agentic LLM tool-use plugins, GitOps controllers, K8s admission, identity providers, federated login, or any place a `redirect_uri`, `state`, `code`, `access_token`, `id_token`, `assertion`, `client_id`, `client_secret`, `code_verifier`, `code_challenge`, `nonce`, `aud`, `iss`, `sub`, `kid`, or `jku` parameter is processed.

---

name: hunt-rce

description: Hunting skill for remote code execution. Built from 1,218 public RCE bug bounty reports across HackerOne, Project Zero, Intigriti, GitHub Security Advisories, and curated awesome-* lists, plus 2024-2026 meta verified against NVD — React Server Components (CVE-2025-55182), runc Leaky Vessels (CVE-2024-21626), BentoML pickle (CVE-2025-27520/32375), LangChain REPL (CVE-2025-68613), Tekton git resolver (CVE-2026-40938), OpenProject git argument injection (CVE-2026-24685), ingress-nginx (CVE-2025-1974). Use when hunting RCE on web apps, OSS libraries, admin panels, file processors, CI/CD runners, GitOps controllers, K8s admission webhooks, container runtimes, ML inference servers, agentic LLM tools, or anywhere user input reaches a shell, deserializer, template engine, dynamic loader, or git/curl/tar CLI.

---

name: hunt-xss

description: Hunting skill for Cross-Site Scripting (XSS) — DOM-based, stored, reflected, mutation-based (mXSS), and modern variants. Built from public bug bounty reports across HackerOne, Intigriti, Bugcrowd, Huntr, and GitHub Security Advisories, plus 2024-2026 meta verified against NVD — DOMPurify nesting mXSS (CVE-2024-47875, GHSA-gx9m-whjm-85jf), DOMPurify depth-bypass family (CVE-2024-45801, GHSA-mmhx-hmjr-r674), DOMPurify mXSS via Re-Contextualization (GHSA-h8r8-wccr-v5f2), Auth0 nextjs-auth0 returnTo (CVE-2025-67716, GHSA-mr6f-h57v-rpj5), React Server Components DoS family (CVE-2025-67779/55184, GHSA-5j59-xgg2-r9c4), markdown-to-jsx iframe XSS (CVE-2024-21535), listmonk stored XSS → Admin ATO (GHSA-jmr4-p576-v565, CVSS 8.0). Use when hunting DOM XSS, stored XSS, reflected XSS, postMessage XSS, mXSS, sanitizer bypass (DOMPurify / sanitize-html / bleach), OAuth redirect_uri XSS, prototype pollution → XSS gadgets, markdown renderer XSS, Server Components content injection, agentic LLM output injection.

---

name: recon-methodology

---

name: report-writing

---

name: sast-methodology

---

name: triage-validation

---

name: vuln-classes