Skip to main content
skill CX Codex

dast-devils-advocate — Claude Code skill: validate DAST findings

dast-devils-advocate — a skill that catches inflated vulnerabilities in DAST reports.

no license ★ 1,298 updated 2026-06-12

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add H-mmer/pentest-agents --skill dast-devils-advocate -a claude-code
Open on GitHub

What dast-devils-advocate does

Runs after /validate PASS and before /report, applying a six-step checklist to every finding: reproducibility, data visibility, whether impact matches the claimed severity, program exclusions, and the "So What?" test—can an attacker extract concrete value or is it purely theoretical.

Two modes ship inside: single-finding adversarial review (default) and exhaustion-claim validation. The skill doesn't kill findings arbitrarily—it downgrades severity when evidence is weaker than the claim (XSS without browser verification, IDOR returning public data, SSRF with DNS callback only). If a bug turns out worse than suspected, it upgrades the rating.

Core function: reproduce the PoC curl, verify the response actually differs, check if the target patched since discovery. XSS demands browser verification; IDOR tests whether ID enumeration is possible; SSRF attempts to reach cloud metadata. Output is JSON with verdict (DOWNGRADE, KILL, BLOCK, UPGRADE, SURVIVES), each check's result, and one concrete sentence on actual impact.

What it helps with

  • separate verified vulnerabilities from hunter assumptions
  • detect whether the target patched since findings were reported
  • lower a finding's rating when evidence is weaker than the claim
  • confirm data is not already public before calling it leaked
  • verify SSRF actually retrieves data, not just DNS callbacks

Example prompts for the agent

  • Review this IDOR claim: can the attacker really read another user's data, or just publicly visible names?
  • subtype: finding—reproduce the XSS in parameter q and tell me if CSP blocks actual cookie theft
  • subtype: exhaustion—I finished my hunt, verify I actually covered all routes or missed endpoints

How to install dast-devils-advocate

1Skills CLI (Vercel)

npx skills add H-mmer/pentest-agents --skill dast-devils-advocate -a claude-code

2Manually: put the files in place

~/.claude/skills/agent-dast-devils-advocate/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/agent-dast-devils-advocate/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about dast-devils-advocate

How do I install dast-devils-advocate in Claude Code?

Run npx skills add H-mmer/pentest-agents --skill agent-dast-devils-advocate -a claude-code, then open ~/.claude/skills/agent-dast-devils-advocate/SKILL.md. For other platforms: Cursor uses -a cursor, GitHub Copilot uses gh skill install H-mmer/pentest-agents agent-dast-devils-advocate, Gemini CLI uses -a gemini-cli.

What does dast-devils-advocate do?

It audits DAST findings for inflation. It reproduces each bug, checks the evidence, recalibrates severity, and flags theoretical vulnerabilities that do not actually work.

What is the difference between DOWNGRADE and KILL?

KILL means the bug does not reproduce or is not a real vulnerability. DOWNGRADE lowers the rating: the bug is real, but severity is overstated due to constraints (e.g., HttpOnly blocks cookie theft, but XSS remains valid).

Similar skills

All skills for Security →

From the same set43

The whole set →
sast-danger-mapper sast-danger-mapper — a subagent that finds dangerous operations in source files: memory handling, type casts, arithmetic at trust boundaries, resource deallocation. CX Codex +1 ★ 1,298repo stars report-writer The report-writer subagent for Claude Code compiles vulnerability findings into reports for pentests, bounty submissions, and technical writeups. CX Codex +1 ★ 1,298repo stars sast-gap-analyzer sast-gap-analyzer analyzes traced data flows to uncover validation gaps in source code. Catches incomplete boundary checks, type mismatches, and race conditions before risky operations. CX Codex +1 ★ 1,298repo stars sast-gap-analyzer sast-gap-analyzer — a subagent for Claude Code, Cursor and other IDEs that identifies where validation chains break down — missing, insufficient or bypassable checks. Detects bugs between code fragments that each look correct in isolation. CX Codex +1 ★ 1,298repo stars dast-devils-advocate dast-devils-advocate — an adversarial subagent that questions DAST findings, reruns PoC commands, flags publicly accessible data, and lowers inflated severity ratings before you submit reports. CX Codex +1 ★ 1,298repo stars ssti-hunter ssti-hunter is a Claude Code skill that investigates server-side template injection (SSTI). It activates for user-controlled templates, rule engines, and rendered messages. CX Codex +1 ★ 1,298repo stars pentest-agents-hunting The pentest-agents-hunting rule structures security hunting around scope, practical impact, and attack chains. It helps an agent drop weak leads and investigate confirmed bugs more deeply. CX Codex +1 ★ 1,298repo stars ssti-hunter ssti-hunter — a subagent for finding and exploiting Server-Side Template Injection (SSTI) across Jinja2, Twig, Velocity, FreeMarker, ERB, and Handlebars. Maps sandbox blocklists, identifies engine type, and chains bypasses to RCE. CX Codex +1 ★ 1,298repo stars

Project sponsors

Companies that keep this analytics open