The Agent Skills Directory

[SAFE]: The skill consists of instructional materials and local automation scripts that follow standard development practices. All external references point to established and trusted technical documentation sites such as MDN, W3C, and WebAIM.
[COMMAND_EXECUTION]: The included bash scripts (audit_accessibility.sh, generate_component.sh, setup_design_system.sh) are used for local file manipulation and static analysis. They utilize standard utilities like mkdir, cat, and sed for scaffolding tasks and do not perform network requests or access sensitive system credentials.
[INDIRECT_PROMPT_INJECTION]: The skill allows the agent to generate project files based on user-provided inputs. 1. Ingestion points: User-defined component names, visual features, and output directories. 2. Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the skill frontmatter. 3. Capability inventory: The skill has the capability to create directories and write files via the included scripts. 4. Sanitization: The provided bash scripts do not perform input validation or sanitization on user-supplied variables. While this creates a vulnerability surface, it is consistent with the intended purpose of a developer scaffolding tool.

frontend-designer