The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill installs the @membranehq/cli global NPM package and uses npx to fetch the latest version. These resources are provided by the vendor to facilitate the integration.
[COMMAND_EXECUTION]: The skill utilizes the membrane command-line tool to perform actions such as searching for connectors, managing connections, and executing Zixflow API requests.
[PROMPT_INJECTION]: The skill contains an indirect prompt injection surface (Category 8) as it processes data retrieved from Zixflow.
Ingestion points: Data from Zixflow enters the context through membrane action run and membrane request output.
Boundary markers: No explicit instructions or delimiters are used to wrap or sanitize the retrieved content.
Capability inventory: The agent is authorized to execute shell commands via the CLI and perform network operations.
Sanitization: There is no documentation of validation or sanitization for the data returned by the API.
[SAFE]: No malicious or suspicious patterns were detected. The skill correctly uses a centralized authentication system, which prevents the need for handling sensitive API credentials directly within the prompt or local environment.

zixflow