slides-grab-html

Pass

Audited by Gen Agent Trust Hub on Oct 30, 2026

Risk Level: SAFECREDENTIALS_UNSAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [CREDENTIALS_UNSAFE]: The skill instructions direct the agent to utilize a local authentication file located at ~/.codex/auth.json for the 'codex' image provider. Accessing sensitive credential files is a high-risk pattern that could lead to unauthorized access or exposure of credentials.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from multiple sources, creating an attack surface for indirect prompt injection.
  • Ingestion points: The skill reads slide-outline.md (which may contain user-controlled content), imports external templates via slides-grab import-template, and fetches web content through yt-dlp or slides-grab fetch-video.
  • Boundary markers: The instructions do not define delimiters or warnings to prevent the agent from following instructions embedded within these external data sources.
  • Capability inventory: The agent has the capability to execute shell commands (slides-grab), read local files (including the auth file), and write files to the slide directory.
  • Sanitization: There is no mention of sanitizing or validating the content retrieved from external sources before it is processed by the agent.
  • [EXTERNAL_DOWNLOADS]: The skill downloads external resources from the web.
  • It fetches the Chart.js library from JSDelivr (https://cdn.jsdelivr.net/npm/chart.js), which is a well-known and reputable content delivery network.
  • It uses yt-dlp and slides-grab fetch-video to download video assets from external websites into the project directory.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 30, 2026, 06:50 AM