landing-page-generator
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user-provided marketing content and product descriptions to generate landing page components, creating a surface where malicious instructions could potentially influence the agent's output.
- Ingestion points: Product name, tagline, and brand voice materials gathered during the workflow in SKILL.md.
- Boundary markers: Absent. The instructions do not provide explicit delimiters or warnings to the agent regarding instructions embedded in user data.
- Capability inventory: The agent can execute local scripts and generate file system output via scripts/landing_page_scaffolder.py.
- Sanitization: The provided generator script, scripts/landing_page_scaffolder.py, implements robust defensive measures including tsx_string (JSON-based string literal encoding), escape (HTML entity escaping), and tsx_href (protocol-aware URL sanitization).
- [COMMAND_EXECUTION]: The workflow requires the execution of internal utility scripts to perform brand analysis and generate code.
- Evidence: SKILL.md triggers scripts/landing_page_scaffolder.py and marketing-skill/content-production/scripts/brand_voice_analyzer.py.
- Context: These scripts are localized to the skill's infrastructure and perform scoped tasks essential to the skill's primary marketing purpose.
- [DYNAMIC_EXECUTION]: The skill generates Next.js TSX and HTML source code at runtime based on user-supplied configurations.
- Evidence: scripts/landing_page_scaffolder.py is designed to scaffold complete production-ready components.
- Context: This represents the primary function of the skill, and the generator script uses secure string templates and escaping to prevent arbitrary code injection from user inputs into the final source code.
Audit Metadata