secure
Pass
Audited by Gen Agent Trust Hub on Mar 23, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill instructs the agent to enter "AUTONOMOUS MODE" and "Do NOT ask questions," which directs the agent to bypass standard interactive safety loops and user confirmation requirements during the assessment process.- [PROMPT_INJECTION]: Indirect Prompt Injection Risk: The skill processes untrusted local data from project files which creates a surface for indirect prompt injection.
- Ingestion points: The skill reads local project files including dependency manifests (package.json, requirements.txt, etc.), configuration files (.env), and source code.
- Boundary markers: Absent; the skill does not use delimiters or instructions to ignore embedded commands within the scanned files.
- Capability inventory: The agent is authorized to execute shell commands (audit tools) and perform file system reads.
- Sanitization: Absent; the skill does not provide instructions for sanitizing or escaping the content of the files it scans before processing.- [COMMAND_EXECUTION]: The skill directs the agent to execute various system-level audit utilities such as npm audit, pip-audit, and cargo audit based on the detected technology stack. While these are legitimate security tools, they involve shell execution based on project configuration.
Audit Metadata