The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user to install a CLI tool from a remote repository using pip or uv. The source URL https://github.com/sleeplessv/youtube-transcript belongs to the skill's author, representing a vendor-owned resource.
[COMMAND_EXECUTION]: The skill's primary function is to execute the yt-transcript command. It includes capabilities for writing output to files via the -o or --output flags, as well as shell redirection in provided examples (e.g., > transcript.json).
[INDIRECT_PROMPT_INJECTION]: The skill processes YouTube transcripts, which are untrusted external data sources. This creates a surface where malicious instructions hidden in a video's captions could potentially influence the agent's behavior during summarization or analysis.
Ingestion points: Data enters the context via the output of yt-transcript fetch (SKILL.md).
Boundary markers: The instructions do not define specific delimiters or warnings to isolate the fetched transcript text from the rest of the agent's prompt context.
Capability inventory: The skill allows file system writes through CLI flags and shell redirection.
Sanitization: The tool provides a --clean flag designed to strip HTML tags and collapse whitespace, providing a layer of content sanitization.

yt-transcript