The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user to install the @membranehq/cli package via npm. This is a verified vendor package from the organization that authored the skill and is required for the integration to function.
[COMMAND_EXECUTION]: The skill uses membrane CLI commands to perform actions such as membrane action run and membrane request. These commands execute within the vendor's managed environment, providing a secure layer for API interaction.
[DATA_EXFILTRATION]: There is no evidence of unauthorized data transfer. The skill emphasizes using the membrane connect flow for authentication, which keeps credential management server-side and prevents the exposure of raw API keys to the agent or user.
[SAFE]: The skill processes external data from Hubstaff but minimizes risks through platform-level controls.
Ingestion points: Hubstaff API responses (e.g., task lists, activity logs) fetched via membrane action run and referenced in SKILL.md.
Boundary markers: None explicitly present in the skill text.
Capability inventory: The skill is limited to network requests via the Membrane platform; it does not request local file system access or the ability to execute arbitrary local code.
Sanitization: Relies on the Membrane platform's internal handling of structured action parameters and proxy requests.

hubstaff