obsidian

Warn

Audited by Gen Agent Trust Hub on Mar 7, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires installing 'obsidian-cli' via a third-party Homebrew tap (yakitrak/yakitrak/obsidian-cli) which is not among the trusted vendor repositories.
  • [COMMAND_EXECUTION]: The skill utilizes shell commands to execute 'obsidian-cli', allowing the agent to search vault content and perform file system operations like creating, moving, and deleting notes.
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface (Category 8) as it reads and processes data from user-controlled Markdown files.
  • Ingestion points: Reads Markdown notes (*.md) and vault configuration (obsidian.json).
  • Boundary markers: No delimiters or protective instructions are used to distinguish note data from agent commands.
  • Capability inventory: The skill can search, create, rename, and delete files on the local disk.
  • Sanitization: Content read from the vault is processed without validation or filtering.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 7, 2026, 11:53 AM