How do I install the Hunt Ssti skill?

Run npx skillmds@latest add duriantaco/hunt-ssti 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 Ssti 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 Ssti?

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 Ssti free to use?

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

Who published Hunt Ssti?

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

Hunt Ssti by duriantaco — Agent Skill

Name: Hunt Ssti
Author: duriantaco

Hunt Ssti

Investigate authorized server-side template injection boundaries. Use when typed discovery or target-origin behavior identifies server-rendered inputs, evaluated-expression differentials, template-engine errors, rendering workflows, or suspected Jinja, Twig, FreeMarker, Mako, ERB, Smarty, Thymeleaf, or Django template evaluation.

duriantaco d592ceb 2 files · 1.8 KB Updated Sep 17, 2026

File contents

Hunt Template Evaluation Boundaries

Treat this card as advisory prioritization. Reflection and template syntax do not prove server-side evaluation.

Workflow

Preserve a target-observed server-rendered request and its render transition.
Pair a bounded expression with a baseline and same-shape syntax control; change one input.
ssti_fingerprint may escalate to fixed file, environment, and command candidates. Run it only when the complete set is authorized; use command_boundary only on an independent typed sink.
Preserve identity, route, fields, controls, and evidence references. Report validated vulnerabilities independently of any challenge objective.

Evidence Gate

Confirm only a derived expression result absent from the paired control and replayable at the same sink. Reflection, syntax disappearance, engine errors, and framework names remain observations. Keep any contract_missing result suspected rather than promoting it.

Stop Conditions

Stop without a server-rendered replay, reproducible transition, clear control, or policy budget. Do not add arbitrary code or extraction beyond the native set; redact environment data and bodies.

Frequently asked questions