The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the membrane CLI for managing service connections and running actions. These commands are standard for the Membrane ecosystem and necessary for the skill's functionality.
[EXTERNAL_DOWNLOADS]: The skill requires the installation of the @membranehq/cli package from npm. This is an official vendor tool and is considered a safe dependency.
[REMOTE_CODE_EXECUTION]: Logic is executed on the Membrane platform via membrane action run. The skill also allows creating actions dynamically using membrane action create, which generates code based on user descriptions.
[PROMPT_INJECTION]: The skill processes untrusted data from WhosOn chat transcripts, which presents a surface for indirect prompt injection.
Ingestion points: WhosOn chat transcripts and behavioral data entry points in SKILL.md.
Boundary markers: No markers or instructions to ignore embedded commands are present in the skill text.
Capability inventory: Access to the membrane CLI and the ability to run or create actions.
Sanitization: No data sanitization or validation logic is specified in the skill instructions.

whoson