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.
What it is
Axiom MCP is an MCP server that connects AI agents to Axiom's log analytics platform. It enables agents to query logs, traces, and metrics using natural language instead of writing APL (Axiom Processing Language) queries manually. The server translates agent requests into structured queries against your Axiom datasets.
This tool targets SRE teams, DevOps engineers, and developers who use Axiom for observability and want AI assistance for log analysis, incident investigation, and metric exploration. The MCP interface makes Axiom data accessible to any MCP-compatible AI agent.
How it saves time or tokens
Instead of writing APL queries manually during an incident, you describe what you need in natural language. The AI agent constructs the query, executes it via the MCP server, and interprets the results. This reduces the time between 'something is wrong' and 'here is the root cause.' The agent can correlate across multiple datasets without you switching between dashboard views.
How to use
- Add the Axiom MCP server to your agent configuration with your Axiom API token and org ID.
- Start your AI agent (Claude Code, Claude Desktop, etc.) with the MCP server configured.
- Ask natural language questions about your logs, traces, and metrics.
Example
{
"mcpServers": {
"axiom": {
"command": "npx",
"args": ["-y", "axiom-mcp"],
"env": {
"AXIOM_TOKEN": "xaat-your-token",
"AXIOM_ORG_ID": "your-org"
}
}
}
}
Then ask your agent:
- 'Show me error logs from the last hour'
- 'What caused the spike in 500 errors at 3pm?'
- 'Find all requests with latency above 2 seconds'
Related on TokRepo
- Monitoring Tools — Observability and monitoring integrations
- DevOps Tools — Infrastructure and operations tools
Common pitfalls
- The Axiom API token determines what datasets the agent can access. Use a token with read-only permissions scoped to the datasets you want to expose.
- Large time ranges can return massive result sets. Encourage the agent to use narrow time windows or add filters to keep queries efficient.
- The MCP server requires network access to Axiom's API. It does not work in air-gapped environments.
Frequently Asked Questions
Axiom is a cloud-native log analytics and observability platform. It ingests logs, traces, and metrics with no indexing overhead and provides APL (Axiom Processing Language) for querying. The MCP server provides AI agent access to these capabilities.
Any MCP-compatible agent works, including Claude Code, Claude Desktop, and other agents that support the Model Context Protocol. The server runs as a local process that the agent communicates with via stdio.
The AI agent sends queries to Axiom via the MCP server and receives results. The query results pass through the AI model for interpretation. Use scoped API tokens and narrow queries to limit data exposure.
The MCP server is designed for interactive queries, not automated alerting. For automated monitoring, use Axiom's built-in alerting features. The MCP server is best for ad-hoc investigation and incident response.
Axiom uses APL (Axiom Processing Language), which is similar to KQL (Kusto Query Language). The MCP server abstracts this by letting agents ask natural language questions that get translated to APL queries.
Citations (3)
- Axiom MCP GitHub— Axiom MCP server gives AI agents access to log analytics
- MCP Specification— Model Context Protocol for AI agent tool integration
- Axiom Documentation— Axiom observability platform and APL query language
Related on TokRepo
Source & Thanks
Built for Axiom — Modern observability platform.
npx axiom-mcp— Axiom MCP server