skills/firecrawl/openclaw/obsidian/Gen Agent Trust Hub

obsidian

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs the obsidian-cli tool using the Homebrew formula yakitrak/yakitrak/obsidian-cli. This introduces a dependency on a third-party repository not managed by a major platform vendor.
  • [COMMAND_EXECUTION]: The instructions direct the agent to execute shell commands via obsidian-cli to search, create, rename, and delete files within the user's filesystem.
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: The skill reads and searches the content of Markdown files (*.md) within a vault, which may contain untrusted data or hidden instructions.
  • Boundary markers: There are no defined delimiters or warnings instructing the agent to distinguish between user-authored notes and executable instructions.
  • Capability inventory: The agent has the ability to write, move, and delete files based on the processed note content.
  • Sanitization: No specific sanitization or filtering mechanisms are described for handling the contents of the Markdown notes.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 07:53 PM
Security Audit — agent-trust-hub — obsidian