obsidian-bases
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists of instructional content and reference documentation for a declarative YAML configuration system. It does not execute scripts, perform network operations, or access sensitive system files.
- [INDIRECT_PROMPT_INJECTION]: The skill describes a system that processes data from external files (Obsidian notes), creating an attack surface for indirect prompt injection. 1. Ingestion points: Note frontmatter and file metadata are accessed via specific namespaces as described in SKILL.md. 2. Boundary markers: The YAML schema provides structured input boundaries, though the skill does not explicitly address adversarial note content. 3. Capability inventory: The formula DSL supports logic, arithmetic, and HTML rendering. 4. Sanitization: The functions reference in references/FUNCTIONS_REFERENCE.md includes an escapeHTML function to mitigate potential XSS when rendering user-controlled content.
Audit Metadata