NYC

paper-2-web

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS] (SAFE): The skill references standard repositories and tools including GitHub (YuhangChen1/Paper2All, oschwartz10612/poppler-windows) and official software sites like libreoffice.org. These are reputable sources for the intended functionality.
  • [CREDENTIALS_UNSAFE] (SAFE): Documentation properly instructs users to create a .env file for API keys (OpenAI, OpenRouter, Google) and uses generic placeholders (e.g., your_openai_api_key_here). No actual secrets are exposed in the files.
  • [COMMAND_EXECUTION] (SAFE): The installation instructions use standard package managers (apt-get, brew, pip) and invoke the project's own scripts (pipeline_all.py, pipeline_light.py). Commands are transparent and directly related to the skill's primary purpose of document conversion and video generation.
  • [DATA_EXFILTRATION] (SAFE): While the skill makes network requests to AI providers (OpenAI, OpenRouter) and Google (for logo searches), this is consistent with its stated purpose. No patterns suggesting unauthorized data collection or exfiltration were found.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 05:57 PM