obsidian-enterprise-rbac
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill implements a system that reads and acts upon external configuration files, creating a vulnerability surface where data poisoning could influence agent behavior.\n
- Ingestion points: The skill parses
.obsidian-permissionsand.obsidian/community-plugins.jsonfrom the vault filesystem to determine access rules.\n - Boundary markers: No explicit boundary markers or instruction-ignoring delimiters are implemented in the data processing patterns.\n
- Capability inventory: The implementation guide provides patterns for file modification (
vault.modify), file creation (vault.create), and file deletion (vault.delete) based on configuration triggers.\n - Sanitization: Data is processed via standard JSON parsing and regular expression matching for path validation.\n- [COMMAND_EXECUTION]: The skill provides code patterns for automated file operations to enforce security constraints within the Obsidian environment.\n
- Evidence: Use of
this.app.vault.delete(file)andthis.app.vault.modify()to restrict folder access and prevent unauthorized plugin installations or configuration changes.
Audit Metadata