ai-organizer-ui-consolidation

Warn

Audited by Gen Agent Trust Hub on Mar 31, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The instructions require the execution of a shell script located at ./scripts/init-frontend.sh. Since the content of this script is not part of the skill's analyzed text, it represents a risk of arbitrary command execution on the user's system.- [EXTERNAL_DOWNLOADS]: The skill depends on a package.json file for installing Node.js dependencies. This introduces a supply chain risk where malicious or unpinned packages could be introduced during the environment setup phase.- [DATA_EXFILTRATION]: The skill builds a dashboard with access to highly sensitive user information, including Google Drive files, email content, and system status. The consolidation of these sources into a single interface that communicates with various API endpoints requires a rigorous audit to ensure data is not exfiltrated to unauthorized external domains.- [PROMPT_INJECTION]: The skill establishes a surface for indirect prompt injection by processing untrusted data (such as emails, PDFs, and images) through AI models like Gemini.
  • Ingestion points: Files are ingested through the 'FileUploadZone' and retrieved via semantic search from email and local storage.
  • Boundary markers: There are no defined boundary markers or instructions to the agent to disregard commands found within the processed files or emails.
  • Capability inventory: The system possesses capabilities to move, rename, and organize files, synchronize with cloud storage (Google Drive), and manage system background services.
  • Sanitization: The skill documentation does not mention any sanitization, filtering, or validation of data before it is processed by the AI analysis modules.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 31, 2026, 02:03 AM
Security Audit — agent-trust-hub — ai-organizer-ui-consolidation