# Openclaw Model Failover

> OpenClaw plugin: auto-detect rate limits/quota/auth failures and switch sessions to fallback LLMs. Prevents 'API rate limit reached' loops by patching pinned session models.

- Skill: `johnalbertini14-glitch/openclaw-model-failover` (Agent Skill, multi-file: 7 files)
- Install (CLI): `npx skillmds@latest add johnalbertini14-glitch/openclaw-model-failover`
- Raw SKILL.md: https://api.skillmd.com/api/skills/johnalbertini14-glitch/openclaw-model-failover/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: johnalbertini14-glitch (https://skillmd.com/u/johnalbertini14-glitch)
- Updated: 2026-09-09
- Page: https://skillmd.com/skills/johnalbertini14-glitch/openclaw-model-failover

---


# openclaw-model-failover (OpenClaw Plugin)

This is an **OpenClaw Gateway plugin**, not an agent skill.

It mitigates outages by automatically switching the model used for a conversation when it detects:
- Rate limits / quota exhaustion (HTTP 429, RESOURCE_EXHAUSTED)
- Auth/scope failures (e.g. OpenAI: missing `api.responses.write`)

## Install

### Option A: ClawHub

```bash
clawhub install openclaw-model-failover
```

### Option B: From repo (dev)

```bash
openclaw plugins install -l ~/.openclaw/workspace/openclaw-model-failover
openclaw gateway restart
```

## Configure

Set model fallback order (example):

- Anthropic -> OpenAI -> Google

See README for the full config block.