mobile-ui-ux
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and apply configuration overrides from project-level files such as
.ui-ux.md,docs/ui-ux-overrides.md, orDESIGN.md. These files originate from the project being analyzed and are therefore untrusted data sources. A malicious project could include instructions within these files intended to manipulate the agent's behavior, skip specific UI/UX rules, or influence the final rubric scoring. - Ingestion points: Project root files (
.ui-ux.md,docs/ui-ux-overrides.md,DESIGN.md) as defined in the 'Workflow' section ofSKILL.mdand detailed inCUSTOMIZE.md. - Boundary markers: None identified. The instructions direct the agent to load the first matching file and use its content to set project profiles, supported devices, and rule exceptions.
- Capability inventory: The skill utilizes agent capabilities for file system reading, screenshot evaluation, and generating structured feedback and scores.
- Sanitization: No sanitization or validation of the content within the override files is specified, beyond identifying expected YAML keys.
Audit Metadata