The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the installation of the @membranehq/cli package from NPM. This is a utility provided by the skill's vendor to facilitate communication with the Membrane platform.
[COMMAND_EXECUTION]: Utilizes the membrane CLI tool to perform various operations, including authenticating users, searching for API connectors, and executing actions against the Avochato API.
[DATA_EXFILTRATION]: Communicates with Avochato's API endpoints. This interaction is mediated by the Membrane platform, which manages credentials and tokens server-side to prevent local exposure of sensitive keys.
[PROMPT_INJECTION]: Indirect Prompt Injection Risk:
Ingestion points: The skill retrieves untrusted data from external sources via actions like list-messages, get-message, and search-contacts (File: SKILL.md).
Boundary markers: No explicit delimiters or instructions are provided to the agent to treat retrieved message content as untrusted data.
Capability inventory: The skill has the ability to perform state-changing operations such as send-message, close-ticket, and arbitrary API requests via membrane request (File: SKILL.md).
Sanitization: There is no evidence of sanitization or filtering of the text retrieved from the Avochato platform before it is processed by the agent.

avochato