obsidian-bases
Pass
Audited by Gen Agent Trust Hub on Jul 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill manages structured .base configuration files for the Obsidian application. It accesses local vault metadata, such as file paths, sizes, and modification dates, which is required for its intended functionality of building database views.\n- [PROMPT_INJECTION]: The skill processes user-controlled data from Obsidian notes (frontmatter and file content), creating a surface for indirect prompt injection. However, the risk is minimal as the skill's capabilities are limited to generating YAML configurations and it lacks high-risk operations like network exfiltration or system command execution.\n
- Ingestion points: Note frontmatter properties via file.properties and file objects via the file() function.\n
- Boundary markers: Absent. No specific instructions are provided to the agent to distinguish note data from potential malicious instructions.\n
- Capability inventory: File-write operations for creating and editing .base configuration files.\n
- Sanitization: No explicit sanitization requirements are defined in the workflow, although an escapeHTML() function is included in the functions reference.\n- [COMMAND_EXECUTION]: No patterns for shell command execution, subprocess spawning, or unauthorized system access were detected.
Audit Metadata