# Axiom MCP — Log Search and Analytics for AI Agents

> MCP server that gives AI agents access to Axiom log analytics. Query logs, traces, and metrics through natural language for AI-powered observability and incident response.

## Install

Merge the JSON below into your `.mcp.json`:

## Quick Use

```json
{
  "mcpServers": {
    "axiom": {
      "command": "npx",
      "args": ["-y", "axiom-mcp"],
      "env": {
        "AXIOM_TOKEN": "xaat-...",
        "AXIOM_ORG_ID": "your-org"
      }
    }
  }
}
```

Now ask your AI agent:
```
"Show me error logs from the last hour"
"What caused the spike in 500 errors at 3pm?"
"Find all requests with latency > 2 seconds"
"Compare today's error rate with yesterday"
```

## What is Axiom MCP?

Axiom MCP is a Model Context Protocol server that connects AI agents to Axiom's log analytics platform. Agents can query logs, traces, and metrics using natural language — no APL (Axiom Processing Language) knowledge needed. The agent translates questions into APL queries, executes them, and interprets results. Ideal for AI-powered incident response and observability.

**Answer-Ready**: Axiom MCP gives AI agents log search and analytics access. Query logs, traces, and metrics via natural language. Agent auto-generates APL queries. AI-powered incident response and observability. Works with Claude Code and Cursor.

**Best for**: DevOps teams wanting AI-assisted log analysis and incident response. **Works with**: Claude Code, Cursor, Claude Desktop. **Setup time**: Under 3 minutes.

## Core Tools

| Tool | What It Does |
|------|-------------|
| query_logs | Search logs with APL or natural language |
| get_datasets | List available log datasets |
| query_apl | Execute raw APL queries |
| get_stats | Summary statistics for a dataset |

## Use Cases

| Use Case | Example Query |
|----------|--------------|
| Incident Response | "What errors appeared in the last 30 minutes?" |
| Performance | "Average response time for /api/users endpoint today" |
| Debugging | "Show me all requests from user_id 12345" |
| Trending | "Compare error rates: this week vs last week" |
| Anomaly Detection | "Any unusual spikes in 4xx errors?" |

## Example Interactions

### Error Investigation

```
User: "Why are users getting 500 errors?"
Agent: [calls query_logs with APL filter for status=500]
Agent: "I found 47 500 errors in the last hour.
        90% are from the /api/payments endpoint.
        Root cause: ConnectionRefusedError to payment-service:8080.
        The payment service appears to be down."
```

### Performance Analysis

```
User: "Is the API getting slower?"
Agent: [calls query_apl with percentile analysis]
Agent: "P95 latency increased from 180ms to 450ms
        starting at 2:30 PM. The slowdown correlates
        with a deployment at 2:28 PM (commit abc123)."
```

## FAQ

**Q: Do I need to know APL?**
A: No, the AI agent translates natural language to APL. But APL knowledge helps for complex queries.

**Q: What log platforms does it support?**
A: This MCP is specifically for Axiom. For other platforms, look for Datadog MCP or Sentry MCP.

**Q: Can it create alerts?**
A: Currently read-only — query and analyze. Alert creation would need the Axiom API directly.

## Source & Thanks

> Built for [Axiom](https://axiom.co) — Modern observability platform.
>
> `npx axiom-mcp` — Axiom MCP server

<!-- ZH -->

## 快速使用

```json
{"mcpServers": {"axiom": {"command": "npx", "args": ["-y", "axiom-mcp"]}}}
```

让 AI Agent 用自然语言查询日志和指标。

## 什么是 Axiom MCP？

MCP 服务器，连接 AI Agent 到 Axiom 日志分析平台。自然语言查询日志、追踪和指标。

**一句话总结**：Axiom 日志分析 MCP，AI Agent 自然语言查询日志/追踪/指标，自动生成 APL，AI 驱动的事件响应和可观测性。

## 核心工具

query_logs、get_datasets、query_apl、get_stats。

## 常见问题

**Q: 需要会 APL？**
A: 不需要，AI 自动翻译自然语言。

## 来源与致谢

> [axiom.co](https://axiom.co) — 现代可观测性平台

---
Source: https://tokrepo.com/en/workflows/655bef8a-41bb-4eda-8f5b-29d1d4cb8c74
Author: MCP Hub