The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructions include steps to download and install the youtube-pp-cli binary using Node.js (npx) and Go (go install) from the author's official packages and repositories.
[COMMAND_EXECUTION]: The skill operates by executing the youtube-pp-cli binary to interact with the YouTube Data API and manage a local databank.
[DATA_EXFILTRATION]: The CLI tool includes a --deliver webhook:<url> feature, enabling the agent to POST command results to an external endpoint. While a functional feature, it provides a potential mechanism for exfiltrating collected research data.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from YouTube, which could contain malicious instructions designed to influence the agent.
Ingestion points: The skill reads YouTube comments (comments-mine), transcripts (videos-transcript), and metadata (packaging) from external channels.
Boundary markers: No specific boundary markers or instructions to ignore embedded commands are present in the processing logic.
Capability inventory: The skill possesses the ability to execute shell commands, write to the local file system via --deliver file:, and perform network operations via the webhook delivery sink.
Sanitization: There is no evidence of sanitization or filtering applied to the external content before it enters the agent's context.

pp-youtube