The Agent Skills Directory

[DATA_EXFILTRATION]: The skill documents the use of auth login-service, which extracts session cookies from the user's browser to access member-only podcast content. It also includes an output delivery mechanism via --deliver webhook:<url>, which allows for the transmission of processed data to arbitrary external URLs.
[EXTERNAL_DOWNLOADS]: The skill provides instructions to install external binaries using npx from the @mvanhorn/printing-press-library package and go install from github.com/mvanhorn/printing-press-library. These resources are associated with the skill's author.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process long-form podcast transcripts from various external sources (YouTube, Substack, RSS feeds). This represents an attack surface where malicious instructions embedded in transcripts could influence the agent's behavior.
Ingestion points: External podcast URLs, including YouTube videos and Substack posts, via episode get and episode batch commands.
Boundary markers: None explicitly defined in the instructions to prevent the agent from executing instructions found within the transcripts.
Capability inventory: The skill has access to the Bash tool and the podcast-goat-pp-cli binary, which can write files to the local system (--out) and perform network requests (--deliver webhook).
Sanitization: The skill does not describe any sanitization or filtering of the transcript content before delivering it to the agent.
[COMMAND_EXECUTION]: The skill requires the execution of a specialized CLI binary (podcast-goat-pp-cli) and provides recipes that involve executing multiple shell commands and scripts to manage transcripts and authentication.

pp-podcast-goat