The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user or agent to install the @membranehq/cli package from the npm registry and uses npx to execute the latest version of the tool. These are vendor-controlled resources from the Membrane organization.- [COMMAND_EXECUTION]: The skill heavily relies on the membrane command-line interface to perform operations such as logging in, connecting to the service, and running actions. These commands are executed in the local shell environment.- [PROMPT_INJECTION]: The skill processes data from the Medius API, which introduces a surface for indirect prompt injection.
Ingestion points: Data retrieved through membrane action run and membrane request in SKILL.md is processed by the agent.
Boundary markers: Absent. The skill does not implement specific delimiters to isolate external data from instructions.
Capability inventory: The agent can execute arbitrary CLI commands using the membrane tool as described in SKILL.md.
Sanitization: Absent. There is no evidence of data validation or escaping for the content received from the external API.- [SAFE]: The documentation contains a mismatched URL in the 'Official docs' section, pointing to Oculus developer documentation instead of Medius. This appears to be a documentation error rather than a malicious attempt to deceive, as the destination is a well-known and trusted service.

medius