FlowClaw is a high-availability LLM load balancer and usage monitor designed to keep AI agents running by rotating provider accounts based on real-time credit availability.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install flowclaw
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 flowclaw 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 FlowClaw?
FlowClaw acts as an unkillable management layer for AI coding agents, specifically designed to integrate with the Openclaw Skills ecosystem. It provides a unified dashboard and automated routing for multiple LLM subscriptions, including Anthropic Claude, Google Gemini, OpenAI Codex, and local Ollama instances. By monitoring usage resets and quotas in real-time, FlowClaw prevents agent downtime and ensures that no credit goes to waste by prioritizing accounts with expiring quotas.
This skill uses an Earliest Deadline First (EDF) scheduling logic, meaning accounts resetting soonest are prioritized. Whether you are managing multiple professional subscriptions or balancing cloud usage with local models, FlowClaw provides the visibility and automation needed to maintain a seamless development workflow without manual intervention.
FlowClaw Use Cases
- Preventing agent interruptions when hitting provider rate limits during intensive coding sessions.
- Automatically prioritizing accounts with credits expiring soonest to maximize subscription value.
- Failing over to local models like Ollama when cloud services are unreachable or quotas are exhausted.
- Managing dozens of OAuth accounts across different LLM providers seamlessly within Openclaw Skills.
- Monitoring weekly and hourly headroom to avoid unexpected service cutoffs.
How FlowClaw Works
- The system monitors real-time usage data across all authenticated providers including Anthropic, Google, and OpenAI.
- It applies a complex scoring algorithm (EDF) to rank every account by urgency, availability, and deadline proximity.
- The skill reorders OpenClaw model priorities based on these rankings to optimize routing and avoid wasted credits.
- It provides a continuous status dashboard to track credit resets and headroom across 5-hour and 7-day windows.
- Automated cron jobs can be configured to silently update these rankings in the background.
FlowClaw Setup
To get started with FlowClaw in your Openclaw Skills environment, follow these provider-specific setup steps:
# For Anthropic Claude Max
claude login
bash {baseDir}/scripts/save-account.sh
# For Google Gemini CLI
openclaw models auth login --provider google-gemini-cli
# For OpenAI Codex
openclaw models auth login --provider openai-codex
# For GitHub Copilot
openclaw models auth login-github-copilot
# Setup cron automation for periodic optimization (every 30 mins)
*/30 * * * * bash ~/clawd/skills/flowclaw/scripts/flowclaw.sh auto
FlowClaw Data Schema & Taxonomy
FlowClaw organizes usage data and provider metadata into a structured format for the scoring engine. This ensures that Openclaw Skills can make informed routing decisions.
| Attribute | Description |
|---|---|
| Urgency | Calculated as remaining credits divided by hours to reset |
| Availability | The dampened capacity of the account using a square root of remaining units |
| Proximity | The pressure of the deadline based on the time window |
| Weekly Headroom | The 7-day capacity reserve status |
| Tier Bonus | Preference weighting for Free, Subscription, or Local models |
FlowClaw Advanced Features
- Family-aware routing ensures that account swaps only occur between models of the same capability class (e.g., swapping one Opus account for another Opus account).
- Automated background optimization via cron jobs to ensure your Openclaw Skills environment is always using the most efficient provider.
- Comprehensive scoring engine that removes penalties when approaching weekly resets to intentionally burn remaining credits.
- Support for local high-parameter models like Ollama Qwen3 235b for zero-cost failover environments.
- Detailed decision history tracking to audit why specific routing changes were made by the load balancer.
SKILL.md
Loading
Related Openclaw Skills
agent-browser
A high-performance CLI tool that enables AI agents to navigate, interact with, and extract data from websites using a structured reference system.
Seedream Image Generation Skill
A professional AI image generation skill leveraging the Seedream 4.5 API to create, edit, and fuse high-quality visual content through intelligent prompt optimization.
OpenClaw Toolbox
A comprehensive management suite for environment initialization, system maintenance, and automated multi-mode backups for AI agent workspaces.
Goal Heartbeat Engine
A robust autonomous state machine that ensures AI agents maintain continuous productivity and balanced attention across multiple long-term goals.
AI Smart Customer Service
A sophisticated LLM-powered customer support agent capable of automated FAQ resolution, intent recognition, and multi-turn dialogue management.
Data Harvester v2
A professional-grade batch web scraping and market intelligence tool for automated data collection and competitor monitoring.
METADATA
Featured*
