The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill installs the official @membranehq/cli from the npm registry. This is a standard dependency provided by the skill author for interacting with their platform and is considered safe for its intended purpose.
[COMMAND_EXECUTION]: The skill instructs the agent to run membrane CLI commands to manage SMS messaging and authentication. These operations are limited to the scope of the D7 SMS service.
[CREDENTIALS_UNSAFE]: The skill demonstrates good security posture by explicitly advising against local storage of secrets. It uses a centralized authentication flow where tokens are managed server-side by the Membrane platform.
[PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes data from an external API. Ingestion points: Output from membrane action run and membrane request in SKILL.md. Boundary markers: Structural JSON formatting enforced by the --json flag. Capability inventory: Execution of shell commands via the membrane CLI. Sanitization: Relies on the agent's ability to parse structured JSON data.

d7-sms