user-guide-creation
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFE
Full Analysis
- [PRIVILEGE_ESCALATION]: The file references/installation.md includes examples of sudo usage within documentation templates. These are static text examples for user instruction and are not executed by the agent.\n- [COMMAND_EXECUTION]: The file references/installation.md contains example wget commands using a placeholder domain (example.com). These serve as illustrative content for generated documentation and do not represent active network requests.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided content to create documentation. Although it lacks explicit boundary markers for untrusted input, the absence of agent tools or execution capabilities prevents any malicious instructions in the input from being executed. Evidence: Ingestion point (user input for doc generation in SKILL.md), Boundary markers (absent), Capability inventory (none), Sanitization (absent).
Audit Metadata