What problem does it solve?
Before running a multi-provider workflow, you need to know which AI providers are actually reachable, how they authenticate, and what they will cost. This Skill produces a one-screen health summary so you never dispatch work to a provider that is down or misconfigured.
Core Features & Use Cases
- Provider Detection: Runs detection scripts to report availability, auth method (env key, OAuth session, subscription), and the model each provider resolves to today.
- Ranked Problem Flagging: Surfaces expired auth first, then version drift (e.g., Claude Code older than v2.1.219 for Opus 5, Codex CLI older than v0.144.0 for GPT-5.6), then missing optional providers, each with the exact fix command.
- Cost Posture Split: Separates included providers (Claude, agy, copilot, ollama, cursor-agent) from billed ones (codex, perplexity, grok, openrouter, atlascloud, claude-sdk) so you can predict workflow spend.
- Use Case: You start a session on a new machine and a workflow banner shows "(unavailable - skipping)". Run this Skill to get a single table of every provider's status plus at most three ranked fix-it lines.
Quick Start
Ask the assistant to run a provider health check and show which AI providers are available, how they authenticate, and what they cost.