obsidian-bases
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill provides an interface for processing untrusted data from the user's Obsidian vault, which constitutes an indirect prompt injection surface.
- Ingestion points: Frontmatter properties of markdown notes and .base YAML configuration files (SKILL.md).
- Boundary markers: Absent; the documentation does not specify delimiters or instructions to ignore embedded logic within ingested data.
- Capability inventory: File system writes to the Obsidian vault and command execution via the rtk ExecutionProxy interface (SKILL.md).
- Sanitization: No mention of escaping or sanitizing external content used in formulas or rendered via the html() function (FUNCTIONS_REFERENCE.md).
- [COMMAND_EXECUTION]: The skill's global protocols define a standardized interface for terminal command execution using the rtk prefix. This is part of the Agentic Company Framework's deterministic execution model and is used for operational tasks such as running tests (SKILL.md).
Audit Metadata