digest
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes content from a local knowledge vault that may contain instructions embedded in data.
- Ingestion points: Reads the audit log, summary layer entries, and evergreen pages (SKILL.md).
- Boundary markers: None; the instructions do not establish delimiters to separate vault content from agent instructions.
- Capability inventory: File reading and text synthesis; the skill does not use network tools or shell command execution.
- Sanitization: None; there is no validation or filtering of vault content before it is processed.
- [NO_CODE]: The skill consists solely of natural language instructions and configuration metadata; no executable scripts or binaries are provided.
Audit Metadata