The Agent Skills Directory

[SAFE]: The skill contains local utility scripts designed to assist in Roblox project maintenance and quality assurance. The behaviors described (linting, type-checking, and reporting) are standard development practices.
[COMMAND_EXECUTION]: The skill includes shell scripts (analyze-all.sh, flags-report.sh, and plan-board-drift.sh) that utilize standard system utilities such as grep, awk, sed, git, and sort to process project files. These operations are scoped to the local repository and do not exhibit malicious command patterns.
[INDIRECT_PROMPT_INJECTION]: The utility scripts ingest data from project files (Lua source code and Markdown documentation). While this represents a data ingestion surface, the scripts primarily perform text extraction and reporting, which are low-risk activities within the context of developer tooling.
Ingestion points: Files in src/ and docs/plans/ are read via grep and awk.
Boundary markers: None identified.
Capability inventory: bash script execution, file reading, and output to console via standard Unix text tools.
Sanitization: Path normalization is performed in analyze-all.sh using sed to convert absolute paths to relative paths.

roblox-diagnostics-and-tooling