The Agent Skills Directory

[SAFE]: The skill serves a legitimate development purpose and contains no indicators of prompt injection, data exfiltration, or malicious intent.
[COMMAND_EXECUTION]: The skill instructs the agent to create local scripts and add task runner entries to the project's configuration (e.g., package.json). These actions facilitate local execution of prototype code using standard tools like tsx and pnpm within the developer's environment.
[DATA_EXPOSURE]: The instructions include best practices for data handling during prototyping, specifically advising the use of in-memory state or ephemeral local files with clear "PROTOTYPE" labels to prevent accidental persistence or exposure of sensitive production data.

ito-prototype