The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructions provide a command to install the omni-captions-skills package from a custom PyPI registry hosted on the author's GitHub Pages domain (https://lattifai.github.io/pypi/simple/). This is a vendor-provided resource necessary for the skill's advertised functionality.
[COMMAND_EXECUTION]: The skill utilizes the omnicaptions and yt-dlp CLI tools through the Bash tool to perform transcription, media processing, and metadata retrieval.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external content, creating a surface for indirect prompt injection.
Ingestion points: YouTube URLs and local audio/video files (documented in SKILL.md).
Boundary markers: None identified; the skill does not specify delimiters or provide instructions to the agent to ignore commands that may be present in the transcribed text.
Capability inventory: The skill has access to shell command execution (Bash(omnicaptions:*), Bash(yt-dlp:*)) and file system access (Read), as configured in the SKILL.md frontmatter.
Sanitization: There is no evidence of automated content filtering or sanitization of the transcription output before it is returned to the agent for processing.

omnicaptions-transcribe