The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill installs the @membranehq/cli package from the public npm registry. This is the official tool provided by the vendor for managing integrations and handling authentication.
[COMMAND_EXECUTION]: Utilizes shell commands (via the membrane CLI) to perform operations like logging in, connecting services, and running actions. These commands are necessary for the skill's stated purpose and follow standard vendor practices.
[PROMPT_INJECTION]: Indirect prompt injection surface identified. The skill ingests data (action names and descriptions) from an external service (Membrane). While these are part of the integration metadata, they could theoretically contain instructions intended to influence the agent.
Ingestion points: Output from membrane action list and membrane action get commands as described in SKILL.md.
Boundary markers: No delimiters or 'ignore instructions' warnings are present in the suggested command patterns.
Capability inventory: Subprocess execution via membrane action run commands.
Sanitization: No explicit sanitization or validation of the action metadata is documented.

feathery