The Agent Skills Directory

[SAFE]: The skill uses official platform tools to manage integrations and credentials, reducing the risk of accidental exposure or credential theft.
[DATA_EXPOSURE]: The skill facilitates access to sensitive payroll and HR data within Paylocity. This behavior is the intended primary purpose of the skill and is performed through a secure proxy (Membrane) that handles authentication.
[COMMAND_EXECUTION]: Instructions involve the installation and use of the @membranehq/cli global npm package. This is the standard CLI for the Membrane platform authored by the same vendor.
[INDIRECT_PROMPT_INJECTION]: The skill has an inherent attack surface for indirect prompt injection as it processes untrusted data from the Paylocity API.
Ingestion points: Data enters the agent context through the output of membrane action run and membrane request commands described in SKILL.md.
Boundary markers: None identified in the provided instructions to help the agent distinguish between its instructions and retrieved data.
Capability inventory: The skill allows the agent to execute actions that can modify Paylocity data via membrane action run and membrane request (documented in SKILL.md).
Sanitization: No specific sanitization or validation of the retrieved API data is mentioned in the skill instructions.

paylocity