# Mohitbansal321 Ai Usage MCP server

Source: https://glimind.com/server/mcp-registry%2Fio.github.mohitbansal321%2Fai-usage-mcp
Registry id: `mcp-registry/io.github.mohitbansal321/ai-usage-mcp`
Status: unknown

## Install

```json
{
  "mcpServers": {
    "mohitbansal321-ai-usage-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "ai-usage-mcp"
      ]
    }
  }
}
```

Claude Code:

```sh
claude mcp add mohitbansal321-ai-usage-mcp-server -- npx -y ai-usage-mcp
```

## Tools (12)

- `cache_health`: One session's cache breaks: when the cached prefix was lost, why, and cost
- `client_usage`: Per-client (Claude Code vs OpenCode) tokens and cost
- `counterfactual_cost`: These tokens at another model's list rates, beside what they actually cost
- `daily_usage`: Per-day tokens and cost, newest day first
- `model_usage`: Per-model tokens and cost
- `plan_usage`: Pro/Max 5-hour and weekly limit usage, when each resets, and the pace
- `project_usage`: Per-project tokens and cost, by the directory a turn ran in
- `recent_sessions`: Recent sessions with project, client, tokens, cost
- `session_usage`: One session: client, model, duration, token breakdown, cost
- `usage_breakdown`: Two or more dimensions at once -- project × day, model × day, client × day
- `usage_report`: Start here: verdict, where usage went, heaviest sessions, what to change
- `usage_summary`: Totals for a period, split by client, tokens + cost

Measured by Glimind (https://glimind.com): MCP handshake + tool list only, no tool is ever called. CC BY 4.0.
