rds-component-mapper

Pass

Audited by Gen Agent Trust Hub on Mar 17, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill instructions define patterns where content from external JSON files (located in 'assets/content/') is rendered using 'v-html' in Vue templates. This creates an indirect prompt injection surface if the JSON content is sourced from untrusted or attacker-controlled origins.
  • Ingestion points: 'assets/content/home.json', 'assets/content/requestinfo.json', and other local content data files (SKILL.md).
  • Boundary markers: None defined in the mapping logic or code snippets.
  • Capability inventory: Mapping natural language to UI components and assigning properties for rendering.
  • Sanitization: The documentation explicitly advises developers to sanitize content if it is user-generated (SKILL.md, Section 5).
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions for installing dependencies from a private registry ('https://npm.edpl.us') and references the '@rds-vue-ui/analytics-gs-composable' for event tracking via Google Analytics, a well-known service.
  • [NO_CODE]: No executable script files were provided in the skill payload; the skill consists entirely of instructional and descriptive markdown files.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 17, 2026, 10:25 PM