money
Pass
Audited by Gen Agent Trust Hub on Aug 10, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns, suspicious remote dependencies, or unauthorized network communications were detected. The skill's behavior is restricted to local data analysis.
- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface due to the ingestion of vault data without sanitization or boundary markers.
- Ingestion points: Obsidian vault files and search results (SKILL.md).
- Boundary markers: Absent; no delimiters are used to wrap retrieved content.
- Capability inventory: Local search tools and text document generation.
- Sanitization: Absent; the agent uses note content directly as evidence for recommendations.
Audit Metadata