obsidian-layout-adjustment
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions mention shell commands such as
screencaptureand AppleScript integration to verify visual layout changes and confirm the application's active state. These operations are limited to local UI verification as part of the styling workflow. - [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing user-supplied styling requests and reading configuration from the local vault's
.obsidian/appearance.jsonfile. 1. Ingestion points: User styling prompts and.obsidian/appearance.json. 2. Boundary markers: The skill emphasizes mapping natural language phrases to stable backend selectors to reduce ambiguity. 3. Capability inventory: Modification of local CSS files in.obsidian/snippets/and execution of local screenshot tools. 4. Sanitization: The skill does not describe explicit sanitization, but relies on the local nature of the vault files. - [SAFE]: No remote code execution, data exfiltration, or obfuscation techniques were identified. The skill's operations are confined to standard Obsidian vault maintenance and CSS styling tasks within the user's local environment.
Audit Metadata