The Agent Skills Directory

[SAFE]: The skill is primarily instructional, providing templates and best practices for documenting Scratch projects. No network exfiltration, credential theft, or unauthorized persistence mechanisms were detected.
[COMMAND_EXECUTION]: The skill suggests using standard local utilities such as unzip and JSON inspection tools to verify the contents of Scratch .sb3 archives. These operations are intended for local development and metadata verification.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it is designed to ingest and process data from external Scratch projects.
Ingestion points: Scratch project archives (.sb3) and their internal project.json metadata files, as noted in the extraction path sections of SKILL.md and references/docstrings.md.
Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded commands within the Scratch project data.
Capability inventory: The skill's capabilities are limited to reading, unzipping, and describing project logic. No high-privilege subprocesses or network capabilities are associated with the processing of this data.
Sanitization: There is no explicit mention of sanitizing or validating the input data from the Scratch files before processing.

scratch-docstrings