The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the installation of the Membrane CLI package (@membranehq/cli) from the official npm registry. This is the vendor's own tool and is necessary for the skill's stated functionality.
[COMMAND_EXECUTION]: The skill uses shell commands via the Membrane CLI to manage connections and execute actions. These commands include parameters like --clientName, --intent, and --input which interpolate strings into the CLI arguments. While these are part of the intended integration, agents should ensure inputs are properly escaped to prevent shell injection.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection.
Ingestion points: Data enters the agent context via membrane action run outputs which fetch property and contact information from RentCast (SKILL.md).
Boundary markers: None identified in the provided instructions for delimiting external data from agent instructions.
Capability inventory: The skill can execute shell commands through the Membrane CLI and write to the local file system if configured by the agent harness.
Sanitization: No explicit sanitization or validation of the external API response is performed before it is presented to the agent.

rentcast