code-step-tour
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local shell commands to create temporary directories, run a provided Node.js script for data processing, and open the generated HTML file in the system browser.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted repository files which could contain malicious instructions.
- Ingestion points: The agent reads project repository files as specified in SKILL.md.
- Boundary markers: There are no explicit instructions or delimiters defined to separate untrusted project content from the agent's instructions.
- Capability inventory: The skill includes file writing to /tmp, script execution, and invoking the 'open' command.
- Sanitization: Data is validated against a JSON schema in the Node.js script and rendered using textContent and escaping in the HTML viewer.
Audit Metadata