generate-slides-retro-simple

Fail

Audited by Gen Agent Trust Hub on Aug 18, 2026

Risk Level: HIGHCREDENTIALS_UNSAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [CREDENTIALS_UNSAFE]: The skill instructions explicitly ask the user to generate a GitHub Personal Access Token and paste it into the chat interface (Plak je GitHub token hier in de chat). It then attempts to export this token to an environment variable. This exposes sensitive credentials to the AI service's logs and chat history.
  • [COMMAND_EXECUTION]: The skill uses several shell commands to interact with the local file system and GitHub API:
  • Uses find and xargs to locate project templates on the user's machine.
  • Uses git clone to download the clidev-presentaties repository from GitHub.
  • Executes npm install to install Node.js dependencies.
  • Uses curl and gh api to fetch repository data, commit history, and project board items.
  • [EXTERNAL_DOWNLOADS]: The skill downloads content from external sources:
  • Clones the repository https://github.com/cedanl/clidev-presentaties.git to the user's home directory.
  • Fetches data from api.github.com using the provided GitHub token.
  • Note: These resources originate from the vendor's own GitHub organization (cedanl).
  • [DATA_EXPOSURE]: The skill accesses and processes various data points from the GitHub organization, including:
  • Commit messages and unique contributor usernames.
  • Project board items, including titles, status, and issue bodies.
  • Source code patches (Python and R files) to calculate code metrics.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Aug 18, 2026, 09:51 PM