The Agent Skills Directory

[Indirect Prompt Injection] (LOW): The skill establishes a workflow where the agent ingests and processes untrusted external data (React source code and YAML metadata), creating a surface for indirect prompt injection.
Ingestion points: index.jsx, component.yml.
Boundary markers: Absent; no instructions are provided to the agent to ignore or delimit instructions found within the components being reviewed.
Capability inventory: File system read/write access (create, modify, refactor, and review components).
Sanitization: Absent; no validation or escaping of the component content is mentioned.
[Data Exposure] (LOW): The instructions suggest checking .env files to resolve directory paths. While a common developer practice, directing an agent to access .env files is a risk factor as these files typically contain sensitive credentials. In this specific context of path configuration, the severity is maintained at LOW.
[No Code] (SAFE): The skill consists entirely of markdown instructions and lacks any scripts, binary executables, or automated network operations.

canvas-component-definition