How do I install the Hunt Xss skill?

Run npx skillmds@latest add duriantaco/hunt-xss in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Hunt Xss safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Hunt Xss?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Hunt Xss free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Hunt Xss?

duriantaco (@duriantaco) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Hunt Xss by duriantaco — Agent Skill

Name: Hunt Xss
Author: duriantaco

Hunt Xss

Investigate authorized reflected, stored, and DOM cross-site scripting boundaries. Use when typed discovery identifies controllable HTML or JavaScript contexts, reflection sinks, stored render workflows, DOM source-to-sink flows, browser execution signals, or suspected executable markup injection.

duriantaco 949b6ea 2 files · 1.8 KB Updated Sep 17, 2026

File contents

Hunt Browser Injection Boundaries

Treat this card as advisory prioritization. Reflection, broken markup, and sink names are not proof of browser execution.

Workflow

Map the controllable source, render transition, context, browser identity, and sink.
Use reflection_value_boundary, then xss_context. Use dom_execution only when its external browser lane is available; it is unavailable with managed authentication.
Compare a unique inert marker with a same-shape encoded control. For stored flows, preserve the write/render chain and clean up the marker.
Retain the source, sink, execution observation, control, and evidence references. Report every validated vulnerability without requiring a challenge artifact.

Evidence Gate

Confirm only controlled browser or DOM execution at the in-scope sink, absent from the paired control. Reflection, context guesses, handler text, and static DOM presence remain observations. Keep any contract_missing result suspected rather than promoting it.

Stop Conditions

Stop without a controllable source-to-sink path, accounted browser lane, clear control, or policy budget. Do not target third-party users, exfiltrate data, or leave persistent effects.

Frequently asked questions