validate

Validates bug bounty findings through a 7-question gate and 4-gate checklist before report writing.

1|Updated Aug 21, 2026
One-click install
npx skills add https://github.com/marcboggs/BMAD-AppSec-Orchestrator --skill validate-marcboggs
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: validate
Source: https://github.com/marcboggs/BMAD-AppSec-Orchestrator/tree/main/bughunter/skills/pentest/validate
Command: npx skills add https://github.com/marcboggs/BMAD-AppSec-Orchestrator --skill validate-marcboggs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Bug bounty hunters often waste time writing reports for findings that get rejected as N/A, duplicates, or out-of-scope, hurting their validity ratio and reputation. This Skill runs a structured validation pass on a finding before any report is written, killing weak findings early. ## Core Features & Use Cases - 7-Question Gate: Checks demonstrability, program acceptance, scope, privilege requirements, known behavior, provable impact, and the never-submit list — one wrong answer kills the finding. - 4 Pre-Submission Gates: Verifies real HTTP reproduction, concrete attacker impact, deduplication against disclosed reports, and report quality (title formula, CVSS, fix recommendation). - Chain Evaluation: Determines whether a conditionally valid finding (open redirect, DNS-only SSRF, clickjacking) can be chained into a reportable impact. - Use Case: After confirming a possible IDOR on an endpoint, run /validate to confirm the response actually exposes another user's private data, check Hacktivity for duplicates, and get a PASS before writing the report. ## Quick Start Run /validate and describe your finding including the endpoint, bug class, what the PoC shows, and the target program to receive a PASS, KILL, or DOWNGRADE verdict.

Frequently Asked Questions about validate

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

FAQPage Schema
How do I validate a bug bounty finding before submitting a report?▼

Run the finding through the 7-Question Gate covering demonstrability, scope, impact, and the never-submit list, then pass the 4 gates for reproduction, impact, deduplication, and report quality. One failed question means kill the finding and move on.

What findings should never be submitted to bug bounty programs?▼

The never-submit list includes missing headers, GraphQL introspection alone, clickjacking without PoC, self-XSS, open redirect alone, DNS-only SSRF, logout CSRF, banner disclosure, and missing cookie flags. These are only reportable if chained into a demonstrated impact.

How do I check if my vulnerability is a duplicate?▼

Search HackerOne Hacktivity for the endpoint and bug class, check GitHub issues, read the five most recent disclosed reports, and review the program changelog for known issues. This dedup check is Gate 2 of the validation process.

When can a weak finding like open redirect still be reported?▼

A weak finding is reportable only when chained into real impact, such as open redirect combined with OAuth code theft for account takeover, or DNS-only SSRF combined with internal service data extraction. Without a confirmed chain, kill the finding.

Why do bug bounty reports get marked as N/A?▼

Reports get marked N/A when the finding lacks demonstrable impact, is out of scope, requires admin privileges an attacker cannot obtain, or is documented known behavior. The validation gate catches these conditions before report writing.