AI Assistant
A bring-your-own-agent framework that investigates traces from the Jaeger UI.
AI Assistant and MCP Server
How Jaeger works with AI agents, inside the UI and from your own tools.
Version2.dev
(preview)
2.23
(latest)
Jaeger exposes trace data to AI agents in two ways, and lets an operator teach
those agents how to analyze it. Both surfaces are served by the jaeger_query
extension on the query port, and both are off until the ai block is
configured.
- The MCP server serves Jaeger’s telemetry tools over the
Model Context Protocol at
/api/ai/mcp/. Any MCP client, such as a coding agent or an IDE, can use it to search traces, map a trace’s structure, compute its critical path, and inspect individual spans. See MCP Server. - The AI Assistant is a bring-your-own-agent framework. A gateway in
jaeger_queryaccepts chat turns over the AG-UI protocol at/api/ai/chat, relays them to an agent of the operator’s choice over the Agent Client Protocol (ACP), and gives the agent the same MCP tools to answer with. The Jaeger UI’s chat panel is one AG-UI client of that gateway; any other AG-UI client can use it directly. See AI Assistant. - Skills are markdown playbooks that teach an agent how to analyze traces. Jaeger ships built-in skills, and an operator can serve their own. See Custom Skills.
Jaeger ships no agent and no model provider, and it holds no model API key. The MCP server works on its own. The assistant works only when an operator runs an agent and points Jaeger at it.
The tools are shaped for progressive disclosure: an agent first searches for candidate traces, then maps a trace without span attributes, then inspects only the spans that look suspicious. A single trace can carry thousands of spans, so this shape keeps a whole trace out of the model’s context window.
Access to either surface is access to your trace data. The security page describes how to put authentication and TLS in front of them and how to limit them to a group of users.
Configuration reference
The ai block of the jaeger_query extension configures both surfaces. Each
field is explained on the page that the row links to.
| Field | Meaning | Page |
|---|---|---|
agent_url | WebSocket URL of the ACP agent; an empty string serves MCP without the assistant | AI Assistant |
agent_headers | Headers sent on every dial to the agent | AI Assistant |
health_check_interval, health_check_timeout | The probe that decides whether the UI shows the chat panel | AI Assistant |
max_request_body_size | Size limit of a chat request | AI Assistant |
auth | Authenticator extension that gates the AI surfaces alone | Security |
mcp | Enables the MCP endpoint; an empty block is enough | MCP Server |
mcp.base_url | Address at which the chat agent reaches the MCP endpoint | AI Assistant |
mcp.skills_dir | Directory of installation-specific skills | Custom Skills |
mcp.streamable_http | Transport options: proxies, replicas, plain JSON responses | MCP Server |
The query extension’s header_forwarding list, outside the ai block, also
reaches the agent; the security page describes it.
Where to Go Next
MCP Server
Jaeger’s trace analysis tools served over the Model Context Protocol for any MCP client.
Custom Skills
Markdown playbooks that teach an agent how to analyze traces, and how Jaeger serves them.
