The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill directs the user to install the @membranehq/cli package from the NPM registry. This is a global installation of the official command-line interface provided by the skill's authoring organization.
[COMMAND_EXECUTION]: The instructions rely on multiple shell commands using the membrane CLI to handle authentication (membrane login), connection management (membrane connect), and service interaction.
[REMOTE_CODE_EXECUTION]: The skill uses the membrane action create command, which utilizes the vendor's platform to dynamically generate and execute logic based on natural language descriptions provided by the agent. This represents a managed form of dynamic code generation.
[PROMPT_INJECTION]: The skill ingests data from external Wire2Air actions, creating an indirect prompt injection surface where external content could potentially influence the agent's behavior.
Ingestion points: Data returned from the membrane action run command in SKILL.md.
Boundary markers: Absent; there are no specific instructions or delimiters provided to help the agent distinguish between trusted instructions and untrusted data from action outputs.
Capability inventory: The agent has access to the membrane CLI, which allows for listing, creating, and executing arbitrary actions on the connected platform.
Sanitization: None observed; the skill instructions direct the agent to process the JSON output of the actions directly.

wire2air