paper-to-html

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill includes external references to the KaTeX library hosted on the official jsDelivr CDN. This is a well-known service used for providing essential math rendering components to the generated HTML analysis.
  • [PROMPT_INJECTION]: The skill utilizes a persona-based prompt for the 'PhD Supervisor Review' section to define a specific tone and structure. While the skill processes external papers which represent a potential indirect prompt injection surface, this risk is managed by the agent's core instructions and the skill shows no malicious intent.
  • [COMMAND_EXECUTION]: The agent is instructed to perform standard file system operations, such as saving paper figures and writing the final HTML analysis file. These are routine tasks for the skill's primary function.
  • [DATA_EXFILTRATION]: No patterns indicative of credential theft or unauthorized data transmission were found. The skill processes input to generate local output without attempting to access sensitive system files.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 08:09 AM