claudian-installer

Warn

Audited by Gen Agent Trust Hub on Apr 13, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONREMOTE_CODE_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses shell commands (mkdir and cp) to create directories and write files to the user's Obsidian plugin folder. These operations modify the local filesystem based on paths that can be influenced by the agent or user input.
  • [REMOTE_CODE_EXECUTION]: The skill's primary purpose is to deploy a Javascript plugin script (main.js) to a location where it will be automatically loaded and executed by the Obsidian application. Because the content of this script is missing from the provided files, it constitutes an unverifiable executable dependency.
  • [REMOTE_CODE_EXECUTION]: According to the provided manifest and description, the plugin being installed possesses 'full agentic capabilities,' including the ability to read and write files and execute bash commands. Installing unverified code with such high-level system access represents a major security surface for potential abuse.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 13, 2026, 03:22 PM