accessibility-audit

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses npm install to download standard libraries including playwright (maintained by Microsoft) and axe-core (a well-known industry standard for accessibility testing).
  • [COMMAND_EXECUTION]: The skill executes a local browser driver script (../dogfood/scripts/browser-driver.mjs) to automate interactions with web pages, capture screenshots, and perform scans.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to navigate to and process content from live web applications, which are untrusted external sources.
  • Ingestion points: Web pages accessed via the Target URL (SKILL.md).
  • Boundary markers: Absent; the skill does not use specific delimiters or instructions to prevent the agent from obeying instructions hidden within the audited web pages.
  • Capability inventory: The skill can execute a local browser driver, write reports to the file system, and capture screenshots.
  • Sanitization: Absent; no specific validation or filtering of the web content is described before it is processed by the agent and axe-core engine.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 01:16 AM
Security Audit — agent-trust-hub — accessibility-audit