# dast-devils-advocate

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

- skill · `type=skill` · Claude Code, Cursor, Codex, GitHub Copilot, Gemini CLI, Windsurf (Devin Desktop), Claude.ai · ★ 1298 repo stars
- Author: H-mmer
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/H-mmer/pentest-agents/tree/HEAD/providers/cursor/.cursor/skills/agent-dast-devils-advocate/SKILL.md
- Updated: 2026-06-12
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/skills/dast-devils-advocate

## What it 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.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add H-mmer/pentest-agents --skill dast-devils-advocate -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/agent-dast-devils-advocate/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/agent-dast-devils-advocate/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add H-mmer/pentest-agents --skill dast-devils-advocate -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/agent-dast-devils-advocate/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add H-mmer/pentest-agents --skill dast-devils-advocate -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/agent-dast-devils-advocate/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install H-mmer/pentest-agents providers/cursor/.cursor/skills/agent-dast-devils-advocate
```

2. **Manually: put the files in place**
- File: `.github/skills/agent-dast-devils-advocate/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add H-mmer/pentest-agents --skill dast-devils-advocate -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/H-mmer/pentest-agents --path providers/cursor/.cursor/skills/agent-dast-devils-advocate
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/agent-dast-devils-advocate/SKILL.md` — The legacy .windsurf/skills/ path still works

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/H-mmer/pentest-agents/tree/HEAD/providers/cursor/.cursor/skills/agent-dast-devils-advocate/SKILL.md

## When to use it
- 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
- 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

## Similar
- [dast-scanning](https://affpapa.org/en/nevibecode/skills/dast-scanning.md) — skill · 411 installs
- [pentesting-checklist-security-assessment](https://affpapa.org/en/nevibecode/skills/pentesting-checklist-security-assessment.md) — skill · 180 installs
- [hunt-race-condition](https://affpapa.org/en/nevibecode/skills/hunt-race-condition.md) — skill · ★ 16193 repo stars
- [triage](https://affpapa.org/en/nevibecode/commands/triage.md) — command · ★ 16193 repo stars
- [subdomain-takeover](https://affpapa.org/en/nevibecode/skills/subdomain-takeover.md) — skill · 4000 installs
- [pentest-agents-bug-bounty-framework](https://affpapa.org/en/nevibecode/skills/pentest-agents-bug-bounty-framework.md) — skill · 1300 installs

From the same set: [h-mmer/pentest-agents](https://affpapa.org/en/nevibecode/by/h-mmer/pentest-agents.md) — 43 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
