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 vendor-provided tool required for the skill's operation and is considered a safe resource from the author.
[COMMAND_EXECUTION]: The skill uses structured CLI commands (membrane login, membrane action run, etc.) to interact with the Membrane platform. These commands are used for legitimate integration purposes and do not involve arbitrary shell execution or unsafe user input interpolation.
[PROMPT_INJECTION]: The skill presents a standard indirect prompt injection surface as it processes action descriptions and data from the Membrane platform.
Ingestion points: Output from membrane action list and membrane action get (SKILL.md).
Boundary markers: Absent.
Capability inventory: Execution of defined actions via membrane action run (SKILL.md).
Sanitization: Authentication and sensitive data handling are managed server-side by the Membrane platform, minimizing local risk.
[SAFE]: The skill adheres to best practices by instructing the agent never to ask for API keys and to use managed connections instead. All external resources and tools identified are official vendor assets.

keka