agent-cost-monitor
Real-time token usage and cost tracking across all your OpenClaw agents — alerts, budgets, and optimization tips
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/bloodandeath/keats-agent-cost-monitorWhat This Skill Does
The Agent Cost Monitor is a critical utility designed for OpenClaw users who need granular visibility into their AI infrastructure. By aggregating data from session logs and token usage metrics, this skill provides a real-time, high-fidelity view of your operational expenditures. It transforms raw usage data into actionable business intelligence by calculating per-agent costs, projecting weekly consumption, and comparing these figures against your predefined budget thresholds. When triggered—either manually via user inquiry or automatically via a scheduled cron job—the skill audits the session_status for all active agents. It flags anomalies such as runaway processes that consume tokens at an unsustainable rate and offers specific, model-aware optimization suggestions, such as recommending a downgrade to a more cost-effective model for specific tasks or adjusting heartbeat frequencies.
Installation
To integrate the Agent Cost Monitor into your OpenClaw environment, execute the following command in your terminal or via the internal CLI:
clawhub install openclaw/skills/skills/bloodandeath/keats-agent-cost-monitor
Ensure that you have sufficient permissions to access the session status logs and that your agent has the required connectivity to query the underlying usage statistics.
Use Cases
This skill is indispensable for power users maintaining fleets of agents. Use it for your weekly cost-hygiene reviews to identify budget leaks, investigate sudden spikes in token consumption when an agent behaves erratically, or automate your financial oversight by scheduling daily reports to be delivered to your dashboard. It is particularly effective for teams running multiple agents on different models (e.g., mixing Opus, Sonnet, and Haiku) where understanding the ROI of each model tier is essential.
Example Prompts
- "Could you generate a detailed cost report for all agents for the last 24 hours?"
- "I'm worried about our spending. Which of my current agents is consuming the most tokens?"
- "Run a budget check; am I likely to exceed my weekly $50 limit at our current usage rate?"
Tips & Limitations
To get the most out of this tool, integrate it into your automated reporting cycle using the provided JSON cron template. Keep in mind that this skill provides estimations based on current provider rates; it is not a direct billing interface. Always check your actual provider dashboard for definitive billing information and invoice management. Avoid using this skill for debugging session crashes, as its primary focus is financial performance, not code execution error recovery.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-bloodandeath-keats-agent-cost-monitor": {
"enabled": true,
"auto_update": true
}
}
}Tags
Flags: data-collection
Related Skills
calling-agent-squad
Activate a multi-agent team (the Squad) to manage complex projects, business tasks, or development workflows. The squad includes a Manager, Architect, Coder, Reviewer, and Observer. Use when the user wants to "call a squad", "start a project", or "deploy squad" with specialized roles and quality control loops.
securityvitals
Security vitals checker for OpenClaw. Scans your installation, scores your setup, and shows you exactly what to fix. First scan in seconds.
feishu-cron-announce
创建定时任务并通过 OpenClaw cron announce 直接推送结果到飞书。当需要设置定时监控任务(如版本发布、天气、股票等)并让结果自动推送到飞书时使用。关键词:飞书定时推送、cron 定时任务、飞书 announce、定时监控、飞书监控通知。
agent-health-diagnostics
Diagnose and fix the 4 most common OpenClaw agent failures — heartbeat spam, API rate limit cascades, channel death loops, and memory/embedding errors. Battle-tested across a 6-agent multi-host deployment.
mempalace-integration
MemPalace记忆系统集成 - AAAK压缩 + Hall分类 + L0-L3分层 30x无损压缩(1000→33 tokens)(1000→33 tokens)(1000→33 tokens)(1000→33 tokens)(1000→33 tokens)(1000→33 tokens),facts/events/preferences/advice分类,加载优先级