obsidian-bases
Pass
Audited by Gen Agent Trust Hub on Jul 21, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of documentation and instructions for generating valid YAML configuration files for an Obsidian plugin. It does not include any executable scripts, remote code downloads, or hidden payloads.
- [DATA_EXPOSURE]: The skill provides documentation for accessing file metadata such as paths, creation times, and frontmatter properties. This is a primary requirement for the skill's purpose—organizing and viewing note data—and does not involve exfiltration to external systems.
- [PROMPT_INJECTION]: There are no patterns suggesting attempts to override system instructions or bypass security constraints. The instructions focus on teaching the agent how to construct valid .base files.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a framework for processing data from notes (frontmatter, tags, metadata) using formulas and filters, which constitutes an ingestion surface for potentially untrusted content.
- Ingestion points: Note properties and file metadata are ingested into formulas defined in SKILL.md.
- Boundary markers: The .base files use YAML structure to delimit instructions from data.
- Capability inventory: The skill is intended for creating and editing configuration files within the Obsidian vault.
- Sanitization: The functions reference includes an escapeHTML() method which can be used to mitigate rendering-related risks.
Audit Metadata