html-comparison-matrix

Pass

Audited by Gen Agent Trust Hub on Jul 7, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns or security risks were identified in the skill's instructions or logic.
  • [DATA_EXFILTRATION]: The skill implements a submission pipeline that sends data back to the agent via a local ephemeral port or the clipboard. This behavior is documented, intended for the skill's functionality, and targets localhost (127.0.0.1), which is a whitelisted environment.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied candidates and criteria. It mitigates potential injection or cross-site scripting (XSS) risks by mandating the use of textContent and document.createElement while explicitly prohibiting the use of innerHTML with variable data.
  • [EXTERNAL_DOWNLOADS]: The skill allows the use of Google Fonts via <link> tags, which is a well-known and trusted service for web artifacts.
  • [COMMAND_EXECUTION]: The skill instructs the agent to run a specific helper tool (html-skills:html-skills-listen) to set up a local receiver. This is a controlled tool invocation within the agent's environment and is necessary for the skill's interactive features.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 7, 2026, 04:18 AM
Security Audit — agent-trust-hub — html-comparison-matrix