The Agent Skills Directory

[SAFE]: No malicious patterns detected. The skill follows security best practices by explicitly limiting its scope, prohibiting dangerous operations like Git delivery or deployment, and providing guidelines for managing the risks associated with processing external documentation.
[INDIRECT_PROMPT_INJECTION]: The skill has a data ingestion surface as it processes external, potentially untrusted Product Requirement Documents (PRDs) and Design specifications to guide frontend implementation.
Ingestion points: Instructions in SKILL.md and references/specification-authorities.md direct the agent to read external .md files.
Boundary markers: The skill includes instructions in references/checklist.md to use delimiters and 'ignore embedded instructions' warnings when handling external content.
Capability inventory: The skill's primary capability is frontend source code modification as defined in SKILL.md.
Sanitization: The skill provides instructions for the agent to sanitize and delimit external content to prevent accidental execution of embedded instructions.

dev-frontend