The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill facilitates the installation of the @membranehq/cli package from the NPM registry. This is a standard dependency for utilizing Membrane-based integrations.
[REMOTE_CODE_EXECUTION]: The instructions involve executing code via npx @membranehq/cli@latest. This pattern dynamically fetches and runs the latest version of the vendor's CLI tool. As this targets a well-known registry and a recognized vendor package, it is considered safe under standard usage but is noted as a remote execution vector.
[INDIRECT_PROMPT_INJECTION]: The skill provides the ability to fetch and process data from Zuplo API endpoints, creating a potential vector for indirect prompt injection if those endpoints return untrusted content.
Ingestion points: Data returned from membrane action run and membrane request commands.
Boundary markers: There are no explicit instructions or delimiters provided to the agent to help distinguish between its instructions and the data fetched from the API.
Capability inventory: The skill allows the agent to perform network operations (membrane request), which could be exploited in a multi-step injection chain.
Sanitization: The skill does not implement or specify any filtering or sanitization of the external API responses before they are processed by the agent.

zuplo