The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: Installs the @membranehq/cli package from the official NPM registry to manage the integration. This tool is an official resource from the skill's authoring organization.
[COMMAND_EXECUTION]: Utilizes the membrane command-line interface to search for connectors, authenticate users, and execute API actions. These operations are standard for the intended integration logic.
[DATA_EXFILTRATION]: Facilitates network communication with the Review Studio API through Membrane's proxy service. This method ensures that sensitive credentials are never exposed to the local environment or the agent directly.
[PROMPT_INJECTION]: The skill handles data from Review Studio, such as annotations and document metadata, which could contain untrusted content. * Ingestion points: Output from membrane action run and membrane request commands in SKILL.md. * Boundary markers: Absent; data is processed as part of the standard workflow. * Capability inventory: The skill uses the membrane CLI for scoped network requests; it does not perform arbitrary file system writes or unauthorized system commands. * Sanitization: No explicit sanitization or filtering of external data is defined in the instructions.

reviewstudio