llmkit-mcp-server
# @f3d1/llmkit-mcp-server
AI cost inspection for supported Claude Code sessions and Cline task data found in VS Code-family
storage. Eleven tools cover local session evidence, authenticated gateway spend, and budget queries.
Part of [LLMKit](https://github.com/smigolsmigol/llmkit), an open-source API gateway with cost tracking and budget enforcement.
## Quick start
Add to your `.mcp.json` (Claude Code) or `.cursor/mcp.json` (Cursor):
```json
{
"mcpServers": {
"llmkit": {
"command": "npx",
"args": ["-y", "@f3d1/llmkit-mcp-server"]
}
}
}
```
The local tools (`llmkit_local_*`) need no API key. They read supported Claude Code sessions and
Cline task data from supported editor storage. Proxy tools require an existing LLMKit API key in
`LLMKIT_API_KEY`. Check [llmkit.sh](https://llmkit.sh) for current account and service availability.
## Tools
### Proxy tools (need API key)
| Tool | What it does |
|------|-------------|
| `llmkit_usage_stats` | Spend, requests, top models for a period |
| `llmkit_cost_query` | Costs grouped by provider, model, session, or day |
| `llmkit_budget_status` | Budget limits and remaining balance |
| `llmkit_session_summary` | Recent sessions with cost, duration, models |
| `llmkit_list_keys` | All keys with status and creation date |
| `llmkit_health` | Proxy ping with response time |
### Local tools (no key needed)
The local tools detect supported installations and aggregate their session data.
| Tool | What it does |
|------|-------------|
| `llmkit_local_session` | Current Claude Code session or latest detected Cline task cost |
| `llmkit_local_projects` | Cumulative cost across all projects and sessions |
| `llmkit_local_cache` | Claude Code prompt caching savings using model-bound prices |
| `llmkit_local_forecast` | 30-day API-rate projection from detected local history |
| `llmkit_local_agents` | Subagent cost attribution (Claude Code) |
### SessionEnd hook
Auto-log session costs when Claude Code exits:
```json
{
"hooks": {
"SessionEnd": [
{
"type": "command",
"command": "npx @f3d1/llmkit-mcp-server --hook"
}
]
}
}
```
## Environment variables
| Variable | Required | Description |
|----------|----------|-------------|
| `LLMKIT_API_KEY` | No | API key for proxy tools. Local tools work without it. |
| `LLMKIT_PROXY_URL` | No | Proxy URL (defaults to hosted service) |
| `LLMKIT_CLINE_DIR` | No | Override Cline data directory path |
| `LLMKIT_SCAN_WSL` | No | Set to `1` to scan WSL homes for Claude Code and Cline data |
## Supported tools
The local tools read data from:
- Claude Code (`~/.claude/projects/`)
- Cline extension storage in VS Code, Insiders, VSCodium, Cursor, and Windsurf
- WSL installations when `LLMKIT_SCAN_WSL=1` (scans distro homes via UNC paths on Windows)
- VS Code, Cursor, and Windsurf server directories for supported remote extension storage
## License
MIT
TDQS
Scored across 11 tools
Tools are mostly distinct, targeting different aspects like budget, cost breakdown, health, keys, local session, etc. However, llmkit_cost_query, llmkit_usage_stats, and llmkit_session_summary have overlapping purposes (cost/spend/usage data) which could cause some confusion.
All tools follow a consistent pattern with the 'llmkit_' prefix and descriptive noun phrases separated by underscores, e.g., llmkit_budget_status, llmkit_cost_query. No mixing of conventions.
With 11 tools, the count is well within the recommended 3-15 range. Each tool covers a distinct aspect of proxy cost monitoring and management, justifying its inclusion.
The tool surface covers major monitoring needs: health, budget, cost, keys, sessions, and local attribution. However, it lacks management capabilities (e.g., create/update keys or budgets) and could include actions like setting usage limits, which are minor gaps for a read-only analytics server.