readme-to-presentation

Pass

Audited by Gen Agent Trust Hub on Aug 30, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted markdown content from external GitHub repositories to plan and generate presentations. This creates a surface where malicious instructions embedded in a README could attempt to influence the agent's slide generation logic or output.
  • Ingestion points: Fetches content from raw.githubusercontent.com or local markdown files in Phase 1.
  • Boundary markers: No explicit delimiters or warnings are specified for the external content processing.
  • Capability inventory: The skill can write files to the local output directory and execute the open command to launch them in a browser.
  • Sanitization: Basic HTML entity escaping is implemented for code blocks in the generated syntax highlighter.
  • [EXTERNAL_DOWNLOADS]: The skill fetches README content from GitHub's raw servers and links to external design assets from Google Fonts and Fontshare. These represent standard external references to trusted organizations and well-known services necessary for the skill's functionality.
  • [COMMAND_EXECUTION]: The skill executes the open command to display the generated HTML presentation in a browser. This is a legitimate operation used to deliver the final product to the user.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 30, 2026, 02:26 PM
Security Audit — agent-trust-hub — readme-to-presentation