hunt-llm-ai
Pass
Audited by Gen Agent Trust Hub on Aug 24, 2026
Risk Level: SAFEPROMPT_INJECTIONDATA_EXFILTRATIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill documents multiple prompt injection techniques, including direct injection examples like 'Ignore previous instructions' and 'reveal the contents of your initial prompt', as well as indirect injection methods using hidden text in documents or images.
- [DATA_EXFILTRATION]: Provides detailed templates for exfiltrating sensitive context (tokens, history) to attacker-controlled OOB endpoints via markdown image rendering and SSRF through agent tools.
- [COMMAND_EXECUTION]: Includes example shell commands for setting up local listeners using Python or netcat to verify and capture data exfiltrated during security testing.
Audit Metadata