The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user to install the Membrane CLI from the official npm registry using npm install -g @membranehq/cli@latest. This is a standard installation procedure for the vendor's tooling.
[COMMAND_EXECUTION]: The skill uses the membrane CLI to perform operations such as login, connection management, and running actions. These commands are scoped to the functionality of the integration and do not exhibit malicious patterns.
[CREDENTIALS_UNSAFE]: The skill explicitly advises against asking users for API keys or tokens, instead utilizing a server-side authentication flow managed by Membrane to handle credentials securely.
[DATA_EXFILTRATION]: While the skill interacts with the Device Magic platform and Membrane's infrastructure, all network operations are consistent with the stated purpose of form automation and data management. There is no evidence of unauthorized data transfer.
[PROMPT_INJECTION]: The skill includes instructions to the agent on how to use the CLI, but does not contain patterns designed to bypass safety filters or override system instructions.
[DATA_INGESTION_SURFACE]: The skill processes data from Device Magic (e.g., form submissions). While this represents an indirect ingestion surface, the risk is mitigated by the structured nature of the data and the use of the Membrane platform as an intermediary.
Ingestion points: Form submissions and resource details fetched via membrane action run.
Boundary markers: None explicitly defined in the instruction templates.
Capability inventory: File system access via CLI installation, network access to vendor domains.
Sanitization: Relies on the Membrane platform's internal handling of API responses.

device-magic