An automated health monitoring and failover utility that ensures continuous AI agent uptime by switching providers during outages.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install model-failover-guard
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install model-failover-guard using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Model Failover Guard?
Model Failover Guard is a mission-critical utility for developers using Openclaw Skills who need to ensure their AI agents remain operational during provider outages or rate-limit issues. It provides a robust monitoring layer that constantly validates the availability of your primary large language model.
By leveraging this skill, you can automate the transition to backup providers without manual intervention. This ensures that your autonomous workflows and integrated applications continue to function seamlessly, making it an essential component for any production-grade implementation of Openclaw Skills.
Model Failover Guard Use Cases
- Maintaining 24/7 uptime for autonomous AI agents.
- Handling API rate limits or service-specific downtime automatically.
- Optimizing cost and performance by managing multiple fallback providers.
- Improving the resilience of complex multi-agent workflows within Openclaw Skills.
How Model Failover Guard Works
- The guard script executes periodic health checks on the designated primary model.
- If the failure count exceeds the defined threshold, the skill selects an available fallback from your configuration.
- The system updates the active model settings to maintain service continuity.
- Monitoring continues until the primary model demonstrates stability over several consecutive checks.
- The skill executes a safe failback, returning the system to its optimal primary state.
Model Failover Guard Setup
To integrate this capability into your project, use the following command:
npx skills add BovmantH/openclaw-model-failover-guard --skill model-failover-guard
Next, configure your monitoring parameters by editing the config.json file located in the skill directory. You can run the guard as a one-time check or a persistent daemon:
# Persistent monitoring
python3 skills/model-failover-guard/scripts/failover.py loop
Model Failover Guard Data Schema & Taxonomy
The skill's behavior is governed by a simple yet powerful JSON configuration schema:
| Key | Definition |
|---|---|
| primaryModel | The default model the agent should use when healthy. |
| failThreshold | Consecutive errors required to trigger a fallback event. |
| recoverThreshold | Successful pings required to confirm the primary model is ready for failback. |
| checkIntervalSec | Frequency of the health monitoring loop in seconds. |
Model Failover Guard Advanced Features
- Supports daemonized background execution for real-time protection.
- Dynamic fallback selection across all models configured within your Openclaw Skills setup.
- Immediate revert logic if a failback attempt detects lingering instability.
- Highly configurable thresholds to balance between sensitivity and stability.
SKILL.md
Loading
Related Openclaw Skills
Ansible Infrastructure Automation
Automate server provisioning, configuration management, and application deployment using Infrastructure as Code.
TiDB Cloud Zero
A specialized tool for provisioning ephemeral, disposable TiDB Cloud Zero databases designed for automated agent workflows and rapid testing.
LocalClaws Meetup Coordinator
A privacy-first agent skill designed to manage local meetup logistics for hosts and attendees via the LocalClaws protocol.
Context Anchor
Context Anchor is a lightweight Bash-based utility that helps AI agents recover from context compaction by scanning memory files to generate a briefing of where they left off.
OpenClaw LLM Router
A unified gateway providing access to over 70 leading LLMs through a single OpenAI-compatible API.
OpenClaw Media Gen
A powerful multimedia generation skill that unifies Gemini image and Qwen Wan video creation through a single API key.
Featured*
