# LLMKit

Track what your AI agents cost across 11 providers. Budget enforcement that actually rejects requests when exceeded. Cost breakdowns by model, session, or provider.

## Quick Start

```bash
# Connect this server (installs CLI if needed)
npx -y smithery mcp add smigolsmigol/llmkit

# Browse available tools
npx -y smithery tool list smigolsmigol/llmkit

# Get full schema for a tool
npx -y smithery tool get smigolsmigol/llmkit llmkit_usage_stats

# Call a tool
npx -y smithery tool call smigolsmigol/llmkit llmkit_usage_stats '{}'
```

## Direct MCP Connection

Endpoint: `https://llmkit--smigolsmigol.run.tools`

**Required config:**
- `apiKey` (query) — LLMKit API key from dashboard

## Tools (6)

- `llmkit_usage_stats` — Get usage statistics (spend, requests, top models)
- `llmkit_cost_query` — Query cost breakdown by provider, model, session, or day
- `llmkit_list_keys` — List all API keys for the account
- `llmkit_budget_status` — Check budget usage and remaining balance
- `llmkit_health` — Check LLMKit proxy health
- `llmkit_session_summary` — Get summary of recent sessions or a specific session

```bash
# Get full input/output schema for a tool
npx -y smithery tool get smigolsmigol/llmkit <tool-name>
```