The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Membrane CLI to perform tasks such as authentication and API interaction.
Evidence: Commands such as membrane login, membrane action run, and membrane request are used to interface with the platform.
[EXTERNAL_DOWNLOADS]: The skill recommends installing the official developer tool for the platform.
Evidence: npm install -g @membranehq/cli is required to use the integration.
[DATA_EXFILTRATION]: No attempts to access sensitive local files or exfiltrate environment variables were found. The skill leverages the vendor's managed connection system to handle credentials securely.
[PROMPT_INJECTION]: The skill processes external data retrieved from the Herald API, creating a surface for indirect prompt injection.
Ingestion points: Data returned from membrane action run and membrane request in SKILL.md.
Boundary markers: Absent; data is returned directly to the agent's context.
Capability inventory: Shell command execution via the membrane CLI.
Sanitization: Absent; the skill relies on the agent's internal safety filters.

herald