The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill recommends installing the @membranehq/cli package. This is an official command-line tool provided by the platform vendor to manage integrations and authentication.
[COMMAND_EXECUTION]: The skill uses the membrane CLI to perform actions such as logging into a tenant, searching for connectors, and running integration actions. These commands are standard operations within the vendor's ecosystem.
[DATA_EXFILTRATION]: Outbound API requests to SecurityTrails are performed via membrane request, which proxies traffic through the vendor's infrastructure. This mechanism prevents sensitive credentials from being handled directly by the agent or stored in local configuration files.
[PROMPT_INJECTION]: The skill has the surface area for indirect prompt injection as it retrieves external data from the SecurityTrails API. * Ingestion points: External records enter the context via membrane action run and membrane request (SKILL.md). * Boundary markers: The instructions do not specify delimiters to distinguish external data from system instructions. * Capability inventory: The agent can execute commands via the membrane CLI. * Sanitization: No explicit validation or filtering of the external API responses is mentioned.

securitytrails