The Agent Skills Directory

[SAFE]: The skill utilizes a centralized collection runner to interface with PostPlus Cloud services. This communication occurs over secure HTTPS or local socket connections to verified vendor endpoints.
Evidence: Found in scripts/collection_actor_run.mjs and the hosted_collection_bridge.mjs library.
[COMMAND_EXECUTION]: Local script execution is restricted to internal utility scripts provided within the skill package. These scripts handle data validation and logging tasks without exposing the system to arbitrary command injection.
Evidence: Execution logic in scripts/collection_actor_run.mjs is tightly controlled.
[EXTERNAL_DOWNLOADS]: The skill includes a utility for downloading videos via the standard yt-dlp library. This is used as intended for research purposes and does not involve untrusted remote code execution.
Evidence: _postplus_shared/00-core/shared-runtime/scripts/download_videos_from_manifest_with_ytdlp.mjs.
[SAFE]: Sensitive information such as CLI session tokens are managed following standard security patterns, using system-appropriate configuration paths and secure network request headers.
Evidence: Configuration management in postplus_cli_config.mjs.

google-trends-research