The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user to install the @membranehq/cli package globally via npm install -g @membranehq/cli@latest. This package is the official command-line tool for the Membrane platform.
[REMOTE_CODE_EXECUTION]: The skill uses the membrane action create command to dynamically generate new executable logic from natural language descriptions. This generated code is subsequently executed via the membrane action run command.
[COMMAND_EXECUTION]: The agent executes various shell commands using the membrane CLI to manage authentication, establish connections to Data247, and perform data operations.
[PROMPT_INJECTION]: The skill provides an attack surface for indirect prompt injection by taking natural language "intents" and "descriptions" and passing them into CLI tools that discover or create executable code.
Ingestion points: User-provided query strings in membrane action list --intent and natural language descriptions in membrane action create.
Boundary markers: None identified to separate the untrusted data from the command structure.
Capability inventory: The CLI can manage authentication, create new code-based actions, and execute these actions with access to the Data247 API.
Sanitization: No sanitization or validation of the input strings is performed before they are processed by the CLI tool.

data247