mcpField
M
model-fallback

model-fallback

github.com/aaaaqwqAgent workflows11159 installsUpdated 1w ago
Repository

About model-fallback

A skill for maintaining model-service continuity when requests fail, time out, or hit limits. It detects failure types, retries requests, and switches among configured backup models and providers, with automatic return to the primary model when it recovers.

Key features

  1. 01Detect request failures and limits
  2. 02Switch to backup models
  3. 03Retry failed requests
  4. 04Fail over across providers
  5. 05Balance load across models
  6. 06Return to primary model after recovery

Use cases

  1. 01Handle model API timeouts
  2. 02Continue service after quota or rate limits
  3. 03Keep requests running when a provider errors

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/aaaaqwq/agi-super-team --skill model-fallback

After installing, type/model-fallbackin Claude Code to confirm it loaded.

Bundled files

References and scripts installed along with SKILL.md.

INSTALL.md
README.md
scripts/
auto-switch-handler.sh
model-error-wrapper.sh
tests/
test-error-handling.sh

Details

From the project README

模型自动降级与故障切换

  • Author: Daniel Li
  • Copyright © Daniel Li. All rights reserved.

当使用此技能

  • 模型请求失败
  • API 超时
  • 达到速率限制
  • 配额耗尽
  • 需要确保服务连续性

工作原理

  1. 监控: 检测模型请求状态
  2. 判断: 识别失败类型（超时/限额/错误）
  3. 切换: 自动降级到备用模型
  4. 重试: 失败请求自动重试
  5. 恢复: 主模型恢复后自动切回

多供应商支持

  • 优先级配置
  • 跨供应商故障转移
  • 智能负载均衡

错误类型

  • 超时: 切换到更快模型
  • 限额: 切换到有配额模型
  • 错误: 切换到不同供应商

触发词

  • "模型降级"
  • "自动切换"
  • "模型故障"
  • "API 失败"

Comments

Markdown supported