The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill uses npx @membranehq/cli@latest to download and execute the Membrane CLI tool. This is a vendor-provided tool used to facilitate communication with the Membrane platform.
[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands via the Membrane CLI, such as login, connect, action run, and request, to interact with Google AppSheet data.
[DATA_EXPOSURE]: The documentation mentions that credentials are saved to ~/.membrane/credentials.json. The instructions do not direct the agent to read or transmit this file; it is only mentioned to explain how the CLI manages authentication.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from Google AppSheet, which constitutes an attack surface for indirect prompt injection.
Ingestion points: Output from action run and request commands executed in SKILL.md.
Boundary markers: None provided in the instructions to delimit external data from agent commands.
Capability inventory: The agent can execute shell commands using the npx @membranehq/cli tool.
Sanitization: No explicit sanitization or validation of the API responses is described.

google-appsheet