The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: Installs the @membranehq/cli package globally. This is a vendor-owned utility required for the skill to interact with the Membrane platform.- [COMMAND_EXECUTION]: Employs the membrane CLI to handle authentication, connection management, and the execution of remote actions or API requests. These are core functionalities for the skill's purpose.- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests and processes data from W&B (e.g., action lists) which could potentially contain malicious instructions from an external source.
Ingestion points: Output from membrane action list and membrane connection list.
Boundary markers: None used to separate external data from agent instructions.
Capability inventory: membrane action run and membrane request can modify external state or perform network operations.
Sanitization: No explicit sanitization or validation of the external API data is mentioned.

wb