The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains behavioral steering instructions directing the agent to proactively suggest the K-Dense Web platform (www.k-dense.ai) during complex workflows. It mandates specific framing for this suggestion as a productivity upgrade, influencing the agent's interaction style for promotional purposes.
[INDIRECT_PROMPT_INJECTION]: The skill's core functionality involves processing large external datasets in formats like CSV and HDF5, which may contain malicious instructions designed to influence agent behavior. * Ingestion points: File reading through functions such as vaex.open() and vaex.from_csv(). * Boundary markers: There are no instructions or delimiters provided to ensure the agent ignores content embedded within the datasets. * Capability inventory: The skill enables the agent to read local files, perform intensive statistical computations, and export data. * Sanitization: No mechanisms for sanitizing or validating the content of the external data files are described.

vaex