The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructions require the agent to read and analyze untrusted external code files, which could contain hidden instructions intended to manipulate the agent's behavior.
Ingestion points: SKILL.md defines a workflow where the agent must "Read the target file(s)" provided for review.
Boundary markers: The instructions do not define specific delimiters or "ignore instructions" wrappers for the content being reviewed.
Capability inventory: The skill is restricted to textual analysis and reporting; it does not utilize tools for network access, file system writes, or command execution.
Sanitization: The skill does not include steps to sanitize or filter the content of the files before processing them.

ui-ux-guidelines