The Agent Skills Directory

[SAFE]: The skill promotes secure credential management by instructing the agent to use Membrane connections, which handle authentication server-side. This avoids the need for the user to provide sensitive API keys or tokens directly to the agent.
[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection because it retrieves and processes data from external AI21 Labs actions.
Ingestion points: Data returned from membrane action run results.
Boundary markers: The instructions do not define specific delimiters or guardrails for the agent when interpreting data fetched from external actions.
Capability inventory: The skill uses the Membrane CLI to execute actions, fetch data, and manage project state.
Sanitization: There is no evidence of explicit content sanitization or validation for the data retrieved from external sources.

ai21-labs