The Agent Skills Directory

Indirect Prompt Injection (HIGH): The skill is designed to ingest untrusted data and has write-access capabilities, creating a significant injection surface.
Ingestion points: Uses Read, Grep, and Glob tools to process React source files as defined in SKILL.md.
Boundary markers: Absent. There are no instructions provided to the agent to distinguish between its own logic and instructions found within analyzed files.
Capability inventory: The skill is granted Write and Edit permissions in the allowed-tools section, allowing it to modify files on the system.
Sanitization: Absent. There is no evidence of content validation or filtering.
Command Execution (MEDIUM): The skill is granted broad file system modification rights (Write, Edit) through its allowed tools. While standard for a 'generator' tool, these capabilities increase the potential impact of a successful injection attack.

store-generator