The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The script _postplus_shared/00-core/shared-runtime/scripts/download_videos_from_manifest_with_ytdlp.mjs fetches media from arbitrary URLs specified in a manifest file using the yt-dlp tool.
[COMMAND_EXECUTION]: The skill runtime utilizes node:child_process to execute external binaries, including python3 (to run the yt-dlp module) and ffmpeg (to generate preview videos), which increases the potential attack surface if inputs are manipulated.
[CREDENTIALS_UNSAFE]: The file _postplus_shared/00-core/shared-runtime/scripts/lib/postplus_cli_config.mjs manages sensitive session tokens (cliSessionToken) by reading and writing them in plain text to the local configuration directory (e.g., ~/.config/postplus/config.json).
[PROMPT_INJECTION]: The skill processes untrusted user-supplied 'briefs' to drive narrative routing decisions. It lacks explicit boundary markers or instructions to ignore embedded commands, creating a surface for indirect prompt injection.
Ingestion points: SKILL.md (brief input) and download_videos_from_manifest_with_ytdlp.mjs (manifest source URLs).
Boundary markers: Absent.
Capability inventory: Shell execution (python3, ffmpeg), network requests (https), and file system writes across the shared runtime.
Sanitization: No explicit sanitization or validation of the brief text content is performed before processing.

pattern-router