accessibility-per-component
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill invokes industry-standard accessibility testing tools using npx (e.g., axe, playwright, test-storybook). Evidence: Found in allowed-tools and Step 1 instructions.
- [COMMAND_EXECUTION]: The skill uses the Bash tool to execute a restricted set of commands for file discovery and running accessibility tests. Evidence: The allowed-tools field explicitly limits Bash to cat, find, head, ls, and specific npx packages.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external component source code and configuration files, which is a potential surface for indirect prompt injection. 1. Ingestion points: Component source files, Storybook URLs, and .ds-ops-config.yml (Step 1). 2. Boundary markers: None explicitly defined for code ingestion. 3. Capability inventory: Read, Write, restricted Bash (SKILL.md). 4. Sanitization: The skill performs analysis on the content rather than executing it as logic.
Audit Metadata