The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user to install @membranehq/cli globally via npm. This is a standard dependency provided by the vendor to facilitate the skill's functionality.
[COMMAND_EXECUTION]: The skill heavily utilizes shell commands through the membrane CLI to manage connections, discover actions, and execute blockchain operations. This is the intended operational mode of the skill.
[DATA_EXFILTRATION]: The skill presents a surface for potential indirect prompt injection (Category 8).
Ingestion points: Data enters the agent context through the output of membrane action run and membrane action list commands executed in the terminal (SKILL.md).
Boundary markers: No explicit instructions or delimiters are provided to the agent to treat action outputs as untrusted data or to ignore embedded instructions within that data.
Capability inventory: The agent has the capability to execute shell commands (membrane action run), create new actions (membrane action create), and perform network-authenticated operations via the CLI (SKILL.md).
Sanitization: The skill does not specify any sanitization or validation logic for the content returned from blockchain actions before it is processed by the agent.

lamden