The Agent Skills Directory

[SAFE]: The skill adheres to security best practices by utilizing a managed authentication architecture and explicitly instructing against the collection or storage of raw API credentials.
[EXTERNAL_DOWNLOADS]: The skill requires the installation of the @membranehq/cli package from npm. This is a legitimate tool provided by the vendor to interface with their platform.
[COMMAND_EXECUTION]: Provides instructions for executing shell commands via the membrane CLI for tasks such as logging in, connecting to services, and querying CRM actions.
[PROMPT_INJECTION]: The skill ingests data from the Kommo CRM, creating a potential surface for indirect prompt injection.
Ingestion points: CRM records (leads, contacts, tasks) retrieved via membrane action run and membrane request as documented in SKILL.md.
Boundary markers: No specific boundary delimiters or instructions to ignore data-embedded commands are provided.
Capability inventory: The agent can perform shell command execution via the membrane tool and execute network requests through the platform's proxy.
Sanitization: No specific data sanitization or validation procedures are implemented within the skill instructions.

kommo