The Agent Skills Directory

[SAFE]: The skill utilizes official tools and services managed by the author (membranedev), ensuring that the infrastructure is consistent with the skill's stated purpose.\n- [EXTERNAL_DOWNLOADS]: The skill installs the Membrane CLI (@membranehq/cli) from the official npm registry, which is a standard procedure for this vendor's ecosystem.\n- [COMMAND_EXECUTION]: The skill uses the membrane command-line tool to perform tasks such as authentication, connection management, and executing Showpad actions. This is the intended operation of the skill and does not involve arbitrary or malicious shell execution.\n- [DATA_EXFILTRATION]: The skill specifically instructs the agent not to handle raw API keys or tokens, instead using server-side connections which reduces the risk of credential exposure or theft.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes data returned from Showpad actions, which represents a potential surface for indirect instructions. However, the use of structured JSON output and the vendor's management of action creation mitigate this risk.\n
Ingestion points: Results from membrane action run and membrane action list commands.\n
Boundary markers: No explicit delimiters are specified in the prompt instructions for handling action output.\n
Capability inventory: Shell execution restricted to the membrane CLI and general network access.\n
Sanitization: Relies on the host environment and the Membrane platform's internal handling of action data.

showpad