workspaces
Pass
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill interacts with external application state, specifically recently opened files and workspace layouts, which represents a potential attack surface for indirect prompt injection.
- Ingestion points: Data enters the agent context through the
obsidian recentsandobsidian workspacecommands. - Boundary markers: The skill does not implement specific delimiters or warnings to ignore instructions embedded within the metadata returned by these commands.
- Capability inventory: The skill is restricted to UI-level interactions via the
obsidianCLI tool, including opening files (tab:open) and switching layouts (workspace:load). - Sanitization: No explicit sanitization or filtering of the output from the Obsidian CLI is defined in the instructions.
Audit Metadata