Skip to main content
subagent CX Codex

dast-devils-advocate — Claude Code subagent: features and setup

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.

no license ★ 1,298 updated 2026-06-12

Install into Claude Code · Manually: put the files in place All methods and platforms ↓
.claude/agents/dast-devils-advocate.md
Open on GitHub

What dast-devils-advocate does

Automated bug hunting tools often exaggerate vulnerability impact. A 200 response code gets labeled as high-severity IDOR, while a reflected string is declared a critical injection. This agent challenges every claim. It replays the original curl commands, queries endpoints without credentials to confirm data is truly private, and consults policy.md and hacktivity.md for program exclusions.

Operating with Bash, Grep, Read, and WebFetch, the agent runs for up to 150 turns. Two modes exist: subtype: finding audits a single vulnerability, while subtype: exhaustion scrutinizes claims that an automated scan finished. Without browser confirmation, client-side scripts get blocked immediately. The subagent then outputs structured JSON with an adjusted title and recalibrated severity.

What it helps with

  • Disprove inflated vulnerability claims prior to reporting
  • Replay proof-of-concept curl requests to verify reproducibility
  • Identify leaked data that is already publicly accessible
  • Recalibrate bug severity based on observed technical impact

Example prompts for the agent

  • Audit the finding in findings/idor-01.json using subtype: finding before generating the final report.
  • Run dast-devils-advocate with subtype: exhaustion to evaluate whether the automated scan thoroughly finished.
  • Review findings/reflected-xss.json and downgrade severity if HttpOnly blocks cookie exfiltration.

How to install dast-devils-advocate

Manually: put the files in place

.claude/agents/dast-devils-advocate.md

For all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention

Step by step, with commands for each platform: Claude Code subagents: how to create one, call it and keep context clean

Questions about dast-devils-advocate

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

Place the agent configuration file at .claude/agents/dast-devils-advocate.md inside your repository.

What does dast-devils-advocate do?

The subagent tests incoming DAST findings adversarially by re-running PoC curls, verifying actual data exposure, and downgrading or rejecting unproven vulnerability claims.

How does the agent handle unverified client-side issues like XSS?

It marks them with a BLOCK verdict until proof is produced from an actual browser, preventing theoretical reflections from being reported as functional exploits.

Similar subagents

All subagents for Security →

From the same set43

The whole set →
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 — a skill that catches inflated vulnerabilities in DAST 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 pentest-agents-techniques The pentest-agents-techniques rule guides agents through checks for authorization flaws and data leaks in web apps. Use it when testing systems you’re authorized to assess. CX Codex +1 ★ 1,298repo stars

Project sponsors

Companies that keep this analytics open