The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill describes methods for processing potentially untrusted external data (such as CMS items and component props) that could be used as a vector for indirect prompt injection.
Ingestion points: The items array in the batchSync function and the data prop in the DataGrid component snippet (SKILL.md).
Boundary markers: The skill does not provide instructions or delimiters to help the agent distinguish between its core instructions and data content processed at runtime.
Capability inventory: The skill utilizes Read, Write, and Edit tools to perform operations on the file system and project files.
Sanitization: The provided TypeScript and React templates demonstrate data transformation for performance but lack explicit sanitization or validation logic for the content being processed.

framer-performance-tuning