webnn

Pass

Audited by Gen Agent Trust Hub on Jun 28, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides legitimate developer resources for implementing the WebNN browser API. Analysis of the provided instructions, references, and scripts indicates no evidence of malicious intent or safety protocol bypasses.
  • [COMMAND_EXECUTION]: The skill utilizes a local utility script scripts/find-webnn-targets.mjs to inventory frontend source files and WebNN markers. The script performs read-only file system operations and does not involve network access or sensitive file exposure.
  • [PROMPT_INJECTION]: The skill scans workspace files to identify integration targets, which constitutes a surface for potential indirect prompt injection if project content is untrusted.
  • Ingestion points: The scripts/find-webnn-targets.mjs script reads content from source files (.js, .ts, .html, etc.) in the workspace.
  • Boundary markers: None. The script matches marker strings directly against the ingested file content.
  • Capability inventory: The skill enables the agent to generate and write integration code based on its findings.
  • Sanitization: None. The script uses simple string matching without content filtering.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 28, 2026, 12:46 AM
Security Audit — agent-trust-hub — webnn