The Agent Skills Directory

[SAFE]: The skill provides guidelines for accessibility auditing without executing code or accessing external resources. Analysis of SKILL.md and the referenced checklists in the references/ directory shows no malicious patterns, command execution, or network activity.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted project source code (UI components, pages, navigation). 1. Ingestion points: Reading UI source code files for Next.js and React Native projects as described in the workflow in SKILL.md. 2. Boundary markers: None specified in the instructions to delimit untrusted code from instructions. 3. Capability inventory: The skill instructions do not invoke any subprocesses, network operations, or file-system write capabilities; it only specifies outputting findings in a report format. 4. Sanitization: No sanitization logic is present for the ingested code content. The risk is minimal because the skill lacks active capabilities to be exploited.

a11y-review