openclaw-usage-mcp
MCP server that reads local OpenClaw session files to provide token usage and cost data without any network calls or authentication.
README
openclaw-usage-mcp
MCP server that reads local OpenClaw session files for usage and cost data. No network calls or authentication required — it scans the JSONL session files directly from disk.
AI Agent ──stdio──▶ MCP Server ──reads──▶ ~/.openclaw/agents/*/sessions/*.jsonl
Tools
| Tool | Description |
|---|---|
get_usage |
Token counts and costs for a date range, broken down by period and model |
list_sessions |
Sessions with usage breakdown, grouped by period (top-N or all) |
get_session_logs |
Per-message logs for a session: timestamps, token counts, costs, roles |
Setup
1. Install dependencies
pip install mcp
2. Configure
The server reads these environment variables:
| Variable | Default | Description |
|---|---|---|
OPENCLAW_AGENTS_DIR |
~/.openclaw/agents/ |
Path to the OpenClaw agents directory containing session files |
3. Register with your MCP client
mcporter (recommended with OpenClaw):
mcporter config add usage \
--command python3 \
--arg /path/to/server.py \
--scope home
Claude Code (~/.claude/settings.json):
{
"mcpServers": {
"openclaw-usage": {
"command": "python3",
"args": ["/path/to/server.py"]
}
}
}
How it works
The server scans JSONL session files from the OpenClaw agents directory:
- Active sessions:
~/.openclaw/agents/*/sessions/*.jsonl - Archived (reset) sessions:
~/.openclaw/agents/*/sessions/*.jsonl.reset.*
Each session file contains JSON lines with type: "session" headers, type: "message" entries (with usage/cost data on assistant messages), and type: "model_change" entries for model switches.
The server aggregates token counts and costs per model, per session, and per time period — returning the same structured output format the tools have always used.
Archived .reset.* files are included in scans, so usage from reset sessions is no longer invisible.
Running standalone
python3 server.py
Communicates over stdio using the MCP protocol.
Previous architecture
Earlier versions used the gateway's WebSocket RPC API with Ed25519 device
authentication. That code (including device_auth.py) is preserved at tag
v0.1.0-gateway-api.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.