The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill recommends installing the @membranehq/cli package from npm. This is an official vendor tool provided by Membrane for managing integrations and authentication.
[COMMAND_EXECUTION]: It utilizes various membrane CLI commands for authentication, service discovery, and API interaction. These commands are fundamental to the skill's functionality and allow the agent to manage its environment.
[PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by ingesting untrusted data through command-line flags like --input and --data in SKILL.md. No boundary markers or sanitization processes are defined. The capability inventory includes the ability to run predefined actions or send proxy requests to the Resurface API as documented in SKILL.md.
[CREDENTIALS_UNSAFE]: The skill follows security best practices by explicitly instructing the agent not to handle raw credentials or API keys, instead utilizing Membrane's built-in connection system for secure authentication.

resurface