agentops-mcp
Exposes analytics from Claude Code transcripts as MCP tools, enabling cost, audit, safety, and efficiency queries through natural language.
README
agentops-mcp
An MCP server that exposes AI-coding-agent operations analytics as tools any MCP client can call. Ask your assistant "how much did I spend on Claude Code this week?", "what dangerous commands did my agent attempt?", or "where are my tokens wasted?" — answered from the transcripts Claude Code already writes locally. Read-only, no telemetry.
Tools
| Tool | Returns |
|---|---|
cost_summary |
tokens, cache hit rate, est. dollars, top projects & tools |
audit_summary |
commands run, file writes, errors, sensitive-path access |
safety_report |
dangerous actions the agent attempted (force-push, rm -rf, curl|sh) |
efficiency_report |
re-read waste, cache misses, concrete fixes |
Each takes days (default 30) and an optional project filter.
Install
pipx install git+https://github.com/Ludoonus/agentops-mcp
# or: uvx --from git+https://github.com/Ludoonus/agentops-mcp agentops-mcp
Use with Claude Desktop (or any MCP client)
Add to your MCP config:
{
"mcpServers": {
"agentops": { "command": "agentops-mcp" }
}
}
Then ask your assistant about your Claude Code cost, audit trail, safety, or efficiency.
How it works
Reads ~/.claude/projects/*.jsonl (the transcripts Claude Code writes), parses the
API-reported usage and tool calls, and serves the analysis over MCP (stdio). Nothing
leaves your machine; the server only reads.
Companion projects
- Operator — the same analytics as a CLI/TUI console.
- cc-powerpack — guardrail hooks that prevent the dangerous actions this server reports.
Go deeper
These tools are the practical layer; The Claude Code Operator's Handbook is the full playbook — 18 chapters on running AI coding agents safely and efficiently (threat model, guardrails, cost, workflows, recovery, scaling). $29, with a free 3-chapter sample.
License
MIT.
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.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
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.