Monitor Claude Code costs, ROI, budgets, anomalies, and usage forecasts.
Copy the install command and let the AI configure it · recommended for beginners
Please install the "io.github.vk0dev/agent-cost-mcp" MCP server from askskill: Run: claude mcp add 'io-github-vk0dev-agent-cost-mcp' -- npx -y @vk0/agent-cost-mcp
Analyze Claude Code tool-call costs from the last 7 days. List total spend, call count, average cost per call, and ROI by tool. Identify the three tools with the highest cost but lowest return, and suggest optimizations.
A cost and ROI report by tool, including a ranked list of high-cost low-return tools and optimization recommendations.
Set a $500 monthly budget for Claude Code usage. Trigger an alert if month-end spending is forecast to exceed the budget by more than 10%, and flag any subagent-tool combinations with abnormal cost spikes today.
Budget monitoring results with overspend forecasts, alert status, and pinpointed abnormal cost sources.
Break down this week's Claude Code costs by subagent. Show which tools each subagent used, how much each cost, what outputs they produced, and assess which subagents are most worth keeping or reducing.
A subagent cost attribution report that helps guide resource allocation and reduction decisions.
Monitor Claude usage live, predict limits, and control costly AI coding actions.
Track AI agent token usage and costs with alerts and task breakdowns.
Attribute multi-tenant LLM usage costs for chargeback billing.
Scan Claude Code configs for vulnerabilities, misconfigurations, and prompt injection risks.
Analyze Claude Code usage, spending, and budgets from a local database.
Check Claude Code rate-limit capacity before launching subtasks.