What problem does it solve? Connecting an agent to an AgentBox platform fails in predictable ways: wrong endpoint, wrong cluster, unreadable output, or writes that silently wait for human approval. This Skill provides the shared operational knowledge every other abx skill relies on for those four concerns. ## Core Features & Use Cases - Authentication setup: Resolve the console endpoint and API key via flags, environment variables, or config file, and verify identity with abx whoami. - Cluster targeting: Choose the correct cluster per command with --cluster, including the single-cluster auto-fill and the --cluster-api exception case. - Machine-readable output: Switch between table, --json, and --csv output, filter lists by column heading, and pipe results into jq or shell loops. - Safe writes with approval flow: Generate write payloads from --schema or --editable instead of memory, and interpret APPROVAL_REQUIRED versus FORBIDDEN_FOR_AGENT responses correctly. - Use Case: An agent's abx envs call fails with an authentication error. Use this Skill to diagnose whether the endpoint, key, or cluster selection is at fault, then re-run with the correct configuration. ## Quick Start Ask the agent to run abx agent-context and then list the environments on a specific cluster using JSON output.
