jaeger-mcp
Read-only MCP server for querying Jaeger environments from Codex, returning compact, sanitized trace summaries to protect sensitive data.
README
jaeger-mcp
Read-only MCP server for querying Jaeger environments from Codex.
The server talks to Jaeger Query API over HTTP and returns compact, sanitized summaries instead of raw traces. This is important because production spans can contain sensitive values such as rpc.request.params, user text, emails, SQL text, headers, and provider payloads.
Environments
This repository does not ship real Jaeger URLs. Configure them locally through environment variables.
Logical environments:
devtestpreviewprod
When two logical environments share the same Jaeger Query endpoint, use grouped URL variables:
export JAEGER_DEV_TEST_URL="<dev-test-jaeger-query-url>"
export JAEGER_PREVIEW_PROD_URL="<preview-prod-jaeger-query-url>"
export JAEGER_DEFAULT_ENV="dev"
export JAEGER_TIMEOUT_SECONDS="20"
JAEGER_DEV_TEST_URL configures both dev and test. JAEGER_PREVIEW_PROD_URL configures both preview and prod.
Explicit per-environment variables override grouped variables:
export JAEGER_DEV_URL="<dev-jaeger-query-url>"
export JAEGER_TEST_URL="<test-jaeger-query-url>"
export JAEGER_PREVIEW_URL="<preview-jaeger-query-url>"
export JAEGER_PROD_URL="<prod-jaeger-query-url>"
For custom environments:
export JAEGER_ENVIRONMENTS='{"dev":"<dev-jaeger-query-url>","prod":"<prod-jaeger-query-url>"}'
For headers:
export JAEGER_HEADERS='{"X-Debug":"1"}'
export JAEGER_PROD_HEADERS='{"Authorization":"Bearer ..."}'
Run
From this repository:
python3 -m jaeger_mcp.server
For Codex MCP configuration, point the command at the source tree without installing:
{
"mcpServers": {
"jaeger": {
"command": "python3",
"args": ["-m", "jaeger_mcp.server"],
"env": {
"PYTHONPATH": "/path/to/jaeger-mcp/src"
}
}
}
}
Replace /path/to/jaeger-mcp with the local checkout path on the machine running Codex.
Tools
jaeger_list_services(env?)jaeger_list_operations(env?, service, span_kind?)jaeger_search_traces(env?, service, operation?, lookback?, start_us?, end_us?, min_duration?, max_duration?, limit?, tags?)jaeger_search_error_traces(env?, service, operation?, lookback?, limit?)jaeger_get_trace_summary(env?, trace_id, max_error_spans?, max_slow_spans?)jaeger_find_error_spans(env?, trace_id, max_spans?, include_logs?)jaeger_get_span_detail(env?, trace_id, span_id, include_logs?)
Design boundaries
- Read-only only.
- Default output is JSON text suitable for Codex reasoning.
- Raw tags are not returned wholesale.
- Sensitive keys are redacted or omitted.
- Long strings are truncated.
- Trace search is capped by
JAEGER_MAX_LIMITwith a default cap of50.
Code layout
server.py: stdio entrypoint.rpc.py: JSON-RPC request handling.tools.py: MCP tool schemas and dispatch.client.py: Jaeger Query API client and tool implementations.trace_analysis.py: span classification, summaries, and trace URLs.redaction.py: sensitive-key detection and output sanitization.config.py: local environment-variable configuration.
Local tests
python3 -B -m unittest discover -s tests
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
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.