The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user to install the Membrane CLI (@membranehq/cli) from the official NPM registry. This is a standard dependency for using the Membrane platform and originates from the vendor's own namespace.
[COMMAND_EXECUTION]: The skill uses the membrane command-line tool to perform actions like logging in, creating connections, and running CRM-specific tasks. These commands are part of the intended functionality for managing the CRM integration.
[DATA_EXFILTRATION]: No evidence of unauthorized data exfiltration was found. The skill explicitly advises letting the platform handle credentials server-side rather than asking the user for API keys, which follows security best practices.
[INDIRECT_PROMPT_INJECTION]: The skill has a potential attack surface as it processes data from external Solve CRM records (like notes or contact details) which could contain malicious instructions. However, the skill utilizes a structured execution environment via the Membrane platform which limits the direct impact of such data on the agent's core logic.
[DYNAMIC_EXECUTION]: The skill includes functionality to create new actions dynamically (membrane action create). While this involves server-side code generation, it is the primary intended purpose of the Membrane platform and occurs within the vendor's managed infrastructure.

solve-crm