material-3-expressive
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill includes a maintenance script,
update_m3_expressive_refs.py, which is configured to retrieve design tokens and UX guidelines from official Google domains, specificallym3.material.ioanddeveloper.android.com. - [COMMAND_EXECUTION]: The folder contains a Python utility script intended for reference maintenance. This script performs authorized network requests and writes updated documentation to the local
references/directory. - [INDIRECT_PROMPT_INJECTION]: The skill exhibits an ingestion surface where external documentation content is used to populate the agent's reference context.
- Ingestion points:
scripts/update_m3_expressive_refs.pyfetches data from remote sources. - Boundary markers: Not present in the generated reference markdown files.
- Capability inventory: The resulting documentation is used by the agent to provide design recommendations.
- Sanitization: The script implements HTML unescaping and regex-based cleaning to sanitize ingested data before saving.
Audit Metadata