managing-obsidian-tasks

Pass

Audited by Gen Agent Trust Hub on Jul 20, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill interfaces with the Obsidian CLI via scripts/obsidian_adapter.py. It uses subprocess.run with argument lists, a secure pattern that prevents shell injection.
  • [PROMPT_INJECTION]: SKILL.md provides strong guidance to the agent, instructing it to use quote-safe transports and avoid raw interpolation of user content into shell commands.
  • [SAFE]: Access is restricted to the user's Obsidian vault and application-specific files like ~/.obsidian-cli.sock, which is consistent with the skill's purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 20, 2026, 09:16 PM
Security Audit — agent-trust-hub — managing-obsidian-tasks