hunt-misc

Guides bug bounty hunting for misc vulnerabilities including access control, SAML, and token scope flaws.

Updated Sep 6, 2026
One-click install
npx skills add https://github.com/inventashif/helpful-code-sidekick --skill hunt-misc-inventashif
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: hunt-misc
Source: https://github.com/inventashif/helpful-code-sidekick/tree/main/scripts/hackerai/skills/bughunter/hunt-misc
Command: npx skills add https://github.com/inventashif/helpful-code-sidekick --skill hunt-misc-inventashif

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Misc vulnerabilities like broken access control, session persistence after removal, SAML signature bypasses, and token scope failures are high-payout but hard to systematically hunt. This Skill provides a structured methodology distilled from 225 public bug bounty reports so hunters can find these bugs consistently instead of relying on intuition. ## Core Features & Use Cases - Attack Surface Mapping: Identifies high-value targets such as role hierarchies, invitation flows, multi-tenant boundaries, internal APIs, and SAML SSO endpoints with concrete URL patterns and header signals. - Step-by-Step Methodology: Provides 12 ordered hunting steps covering invitation bypass, post-removal access, PAT scope fuzzing, cross-tenant access, CRLF injection, SSRF via config fields, and dependency confusion. - Payloads and Validation Gates: Ships ready-to-adapt curl payloads, grep patterns, ReDoS detection commands, a Gate 0 impact validation checklist, and six senior-level vulnerability chains that compose primitives into critical-impact findings. - Use Case: While testing a SaaS platform, you follow the methodology to discover that a removed staff member's session token still accesses customer PII via direct API calls, then use the Body-Diff Rule and Gate 0 questions to write a report that survives triage. ## Quick Start Ask the AI to hunt for misc vulnerabilities on your authorized bug bounty target using the hunt-misc methodology, starting with mapping role and permission boundaries.

Frequently Asked Questions about hunt-misc

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I hunt for access control vulnerabilities on bug bounty programs?

Map every role level from owner to guest, then test whether lower roles can call endpoints meant for higher roles via direct API requests. The methodology requires a response body differential, not just a status code change, to claim a privilege bypass.

How to test SAML SSO for signature bypass vulnerabilities?

Capture a valid SAMLResponse, then try stripping the Signature element, injecting a sibling unsigned Assertion with a victim NameID, or exploiting parser differentials between the signature checker and business logic. These techniques target XML canonicalization and namespace confusion flaws.

Why do bug bounty reports for misc bugs get marked as N/A?

Most rejections come from status-code-only claims without body differences, server-policy blocklists mistaken for file-existence oracles, or missing concrete impact. The Gate 0 validation requires a concrete attacker action, a real victim asset loss, and 10-minute reproducibility.

Can removed users still access a SaaS platform with old session tokens?

Yes, when removal only updates the membership table without invalidating the session store or revoking issued tokens. Test by capturing a session before removal, then replaying API calls afterward to check for continued access to customer data.

What is token scope checked at issuance but not at use?

Some platforms validate PAT or OAuth scopes only when the token is created, while individual API handlers trust middleware and skip re-checking. A read-only token can then perform write actions on endpoints missing scope enforcement.

When should I not report a misc vulnerability finding?

Avoid reporting when you cannot demonstrate concrete impact such as cross-tenant data reads or admin actions, when the behavior is a server-side blocklist rather than a real oracle, or when reproduction requires unrealistic preconditions or luck-based race conditions.