feedback
Pass
Audited by Gen Agent Trust Hub on Oct 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user feedback and data from shared knowledge graphs, which represents a potential attack surface.
- Ingestion points: User-provided feedback strings and data returned from
iskron_semantic_searchqueries (referenced inSKILL.md). - Boundary markers: There are no explicit instructions to use delimiters or ignore-embedded-instructions prompts when handling retrieved content.
- Capability inventory: The skill uses
iskron_caseto write updated information back to shared graphs, along with several search and discovery tools (iskron_search,iskron_orient). - Sanitization: The instructions rely on the model's reasoning to transform input into a specific schema rather than programmatic sanitization.
Audit Metadata