The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or unauthorized data access were detected in the skill instructions.
[COMMAND_EXECUTION]: The skill uses the @membranehq/cli to manage connections and execute actions. This includes membrane action create, which generates integration logic on the server side based on natural language descriptions, and membrane action run to execute those actions.
[EXTERNAL_DOWNLOADS]: The instructions involve installing the @membranehq/cli package from the NPM registry. This is an official tool provided by the skill's vendor for interacting with their platform.
[PROMPT_INJECTION]: The skill processes data from Five°degrees, which introduces a surface for indirect prompt injection if the fetched data contains malicious instructions.
Ingestion points: Data returned from the Five°degrees API via membrane action run as described in SKILL.md.
Boundary markers: Not explicitly defined in the instructions.
Capability inventory: membrane action run (execution of actions) and membrane action create (generation of new logic).
Sanitization: No explicit sanitization or validation of external data is described in the prompt logic.

fivedegrees