plain-plan
Pass
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill ingests user-supplied requirements to generate text and HTML plans. This creates a surface for indirect prompt injection where malicious instructions embedded in the input data could theoretically influence the agent's output or be reflected in the generated HTML content.
- Ingestion points: User requirements for the implementation plan (SKILL.md).
- Boundary markers: Not explicitly defined for the HTML generation process.
- Capability inventory: File writing (HTML to temporary directory) and tool invocation (html-preview).
- Sanitization: The instructions do not specify sanitization or escaping of user input before it is interpolated into the HTML template.
- [COMMAND_EXECUTION]: The skill involves generating local HTML files in the system's temporary directory. While these are functional capabilities for a documentation tool, creating files based on user-controlled input is a behavior that requires careful handling of file paths and content to prevent unintended file system interactions.
Audit Metadata