The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell scripts (tools/grid-lint.sh) and launch macOS application binaries directly from the filesystem (~/Applications/<App>.app/Contents/MacOS/<App>) using environment variables for configuration.
[DYNAMIC_EXECUTION]: The skill utilizes the swift command to dynamically execute a script file (tools/check-layout.swift) against project layout files.
[INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection through its verification workflow:
Ingestion points: The skill processes user-generated or application-generated layout JSON files located in /tmp/<App>-shots/.
Boundary markers: No delimiters or instructions are provided to the agent to treat the content of these JSON files as untrusted data or to ignore embedded instructions.
Capability inventory: The skill environment possesses capabilities to execute shell scripts, run Swift scripts, and launch system binaries as described in SKILL.md.
Sanitization: There is no evidence of sanitization, validation, or escaping of the layout data before it is passed to the check-layout tool or interpreted by the agent.

layout-grid