The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill exposes a customMapFunction parameter described as a "JavaScript function to transform each output object." This allows for the execution of arbitrary code within the Apify actor environment. While this is a feature of the underlying Apify platform, it presents a significant risk if an attacker can inject code into this parameter via indirect prompt injection.
[COMMAND_EXECUTION]: The documentation provides examples of shell command execution using a run_actor.js script to trigger the scraping process. This script handles input parameters and output formatting on the local system.
[EXTERNAL_DOWNLOADS]: The skill uses the Apify API (api.apify.com) and external actors to perform scraping. It also interacts with YouTube to fetch playlist metadata.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from YouTube (video titles, descriptions, and channel info). Because this data is returned to the agent context and can be transformed using the customMapFunction, there is a surface area for instructions embedded in YouTube metadata to influence the agent's behavior or exploit the code execution parameter.
Ingestion points: YouTube playlist URLs, video metadata (titles, descriptions) retrieved via the Apify actor.
Boundary markers: None specified in the provided documentation.
Capability inventory: Subprocess execution via run_actor.js, network access to Apify API, and arbitrary JavaScript execution via customMapFunction.
Sanitization: None mentioned for the external content before processing.

scraping-youtube-playlist