obsidian

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of obsidian-cli via the yakitrak/yakitrak/obsidian-cli Homebrew tap. This is a recognized utility for managing Obsidian vaults from the command line.
  • [COMMAND_EXECUTION]: The skill executes various obsidian-cli commands (e.g., search, create, move, delete) to manipulate files and folders within the user's Obsidian vaults.
  • [DATA_EXPOSURE]: The skill instructions direct the agent to read the Obsidian application configuration file located at ~/Library/Application Support/obsidian/obsidian.json. This access is necessary to programmatically determine vault locations and identify which vault is currently active.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection because it reads and processes the content of Markdown notes which may contain untrusted data.
  • Ingestion points: Note content is ingested into the agent context via the obsidian-cli search-content command.
  • Boundary markers: The instructions do not define specific delimiters or warnings to ignore instructions embedded within the notes.
  • Capability inventory: The agent has the capability to create, move, and delete files within the vault based on its analysis of these notes.
  • Sanitization: There is no explicit sanitization or filtering logic mentioned for the retrieved note content.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 10:13 AM