qt-cpp-docs
Pass
Audited by Gen Agent Trust Hub on May 20, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill includes a 'Guardrails' section that explicitly instructs the agent to treat source content strictly as data and ignore any embedded instructions. This is a recommended safety measure for skills that process user-provided source code.
- [COMMAND_EXECUTION]: The skill uses basic directory listing commands (via
BashorGlob) to verify if documentation already exists. This behavior is limited to the local project structure and involves a user confirmation step (AskUserQuestion) before any files are modified or created. - [DATA_EXFILTRATION]: No network-related commands or exfiltration patterns were found. The skill operates locally by reading source files and writing Markdown documentation to a project-specific
doc/subdirectory. - [INDIRECT_PROMPT_INJECTION]: While the skill processes untrusted source code, it explicitly defines boundary markers and provides clear instructions to the agent to treat the input as technical material only, mitigating the risk of instructions in code comments affecting the agent's operation.
Audit Metadata