accessibility-audit

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill is composed solely of instructional markdown content. It establishes a workflow for accessibility testing (Automated scan, Keyboard inspection, Visual inspection, Semantic inspection) and provides a priority framework for fixing issues. No security risks such as code execution, data exfiltration, or prompt injection were detected.
  • [EXTERNAL_DOWNLOADS]: The skill references an external repository on GitHub as an upstream source for its methodology. This is a standard documentation practice and does not involve any automated code execution or downloading of untrusted payloads.
  • [INDIRECT_PROMPT_INJECTION]: The skill involves analyzing external web pages and UI elements. While this exposes the agent to untrusted data from the web, the skill functions as a guide for evaluation rather than performing unsafe processing of that data. (1) Ingestion points: Web UI and DOM (SKILL.md); (2) Boundary markers: Absent; (3) Capability inventory: No scripts or tool calls defined within the skill itself; (4) Sanitization: Absent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 01:25 PM