hunt-ssti
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEPROMPT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The section titled "Autonomous Testing Priority" instructs the agent to "Escalate straight to RCE — don't stop at arithmetic detection." This encourages the agent to bypass typical security research validation steps and adopt a highly aggressive posture, potentially overriding safer or more conservative operational guidelines.
- [INDIRECT_PROMPT_INJECTION]: The skill directs the agent to ingest and process data from untrusted external web applications, creating a vulnerability surface for indirect injection.
- Ingestion points: The agent is instructed to analyze responses from target endpoints including email templates, PDF/report generators, CMS preview features, and error pages (SKILL.md).
- Boundary markers: None; the instructions do not specify the use of delimiters or directives to ignore instructions that might be embedded in the target application's output.
- Capability inventory: The skill provides the agent with functional OS command execution payloads for multiple environments, including Python/Flask (
popen), PHP/Symfony (exec), and Java/Freemarker (Execute). - Sanitization: The skill lacks instructions for sanitizing, validating, or escaping content returned by the target application before the agent processes it.
Audit Metadata