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 apackage.jsonfile 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