flutter-pub
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves and processes package metadata from pub.dev, which contains user-submitted content. This creates an attack surface for indirect prompt injection.
- Ingestion points: Package information fetched from https://pub.dev/api/packages/ and search results from https://pub.dev/api/search as described in SKILL.md.
- Boundary markers: Absent. The skill instructions do not specify the use of delimiters or 'ignore' commands to isolate external metadata from the agent's internal logic.
- Capability inventory: The skill uses network read capabilities (WebFetch) to access external data.
- Sanitization: Absent. Data is directly interpolated into markdown responses without validation or filtering.
Audit Metadata