The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill installs the @membranehq/cli package from the npm registry. This is the official command-line tool for the Membrane platform, used here to manage the Nexudus integration.
[COMMAND_EXECUTION]: The skill instructs the agent to execute several shell commands using the membrane CLI, such as login, connect, and action list. These commands are used to authenticate the session and interact with the service provider.
[REMOTE_CODE_EXECUTION]: Through the membrane action run command, the skill executes integration logic on Membrane's remote servers. This is the core intended functionality of the skill, allowing it to perform Nexudus operations without local resource overhead.
[PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection because it ingests and processes dynamic metadata (action names and descriptions) from an external service.
Ingestion points: External data enters the context via the output of the membrane action list and membrane action get commands.
Boundary markers: Absent; there are no specific markers or instructions provided to the agent to differentiate between system instructions and data retrieved from the API.
Capability inventory: The skill possesses capabilities for remote action execution (membrane action run) and local command execution.
Sanitization: Absent; the skill does not specify any validation or filtering of the metadata retrieved from the remote service.

nexudus