The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the membrane CLI to interact with Gong. This allows the agent to perform actions like searching for connectors, connecting accounts, and running pre-defined actions.
[EXTERNAL_DOWNLOADS]: Recommends installing the @membranehq/cli tool via npm. This package is provided by the vendor to facilitate secure API communication.
[SAFE]: No hardcoded secrets or sensitive configuration files are accessed. Credential management is handled server-side by the Membrane platform, preventing local credential exposure.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface when handling external data.
Ingestion points: Retrieves call transcripts, meeting details, and user data from Gong (e.g., via get-call-transcripts and get-calls-extensive in SKILL.md).
Boundary markers: Absent; there are no instructions to delimit or ignore instructions found within the retrieved Gong data.
Capability inventory: The skill can perform write operations to Gong, such as create-meeting, update-meeting, and create-call (SKILL.md).
Sanitization: Absent; external data from Gong is processed without explicit sanitization or filtering.

gong