zeroapi

# ZeroAPI — Subscription-Based Model Routing

Route incoming tasks to the optimal AI model across available providers. OpenClaw handles all API connections — this skill defines the classification and delegation logic. Classify each task by type and delegate to the appropriate agent/model.

## First-Time Setup

When this skill is first loaded, determine the user's available providers:

1. Ask: "Which AI subscriptions do you have?" (Claude Max 5x/20x, ChatGPT Plus/Pro, Gemini Advanced, Kimi)
2. Map subscriptions to available tiers (see table below)
3. Disable tiers for missing providers — those decision steps get skipped
4. Confirm the active configuration with the user

If only Claude is available, all tasks stay on Opus. No routing needed — but conflict resolution and collaboration patterns still apply for judging task complexity.

To verify providers are actually working after setup, ask the user to run:
```bash
openclaw models status
```
Any model showing `missing` or `auth_expired` is not usable. Remove it from your active tiers until the user fixes it.

For full provider configuration details, consult `references/provider-config.md` (in the same directory as this SKILL.md).

## Model Tiers

| Tier | Model | OpenClaw ID | Speed | TTFT | Intelligence | Context | Best At |
|------|-------|-------------|-------|------|-------------|---------|---------|
| SIMPLE | Gemini 2.5 Flash-Lite | `google-gemini-cli/gemini-2.5-flash-lite` | 495 tok/s | 0.23s | 21.6 | 1M | Low-latency pings, trivial format tasks |
| FAST | Gemini 3 Flash | `google-gemini-cli/gemini-3-flash-preview` | 206 tok/s | 12.75s | 46.4 | 1M | Instruction following, structured output, heartbeats |
| RESEARCH | Gemini 3 Pro | `google-gemini-cli/gemini-3-pro-preview` | 131 tok/s | 29.59s | 48.4 | 1M | Scientific research, long context analysis |
| CODE | GPT-5.3 Codex | `openai-codex/gpt-5.3-codex` | 113 tok/s | 20.00s | 51.5 | 200K | Code generation, math (99.0) |
| DEEP | Claude Opus 4.6 | `anthropic/clau...