Add Model Context Protocol (MCP) project proposal #3128

pavolloffay · 2025-11-06T10:02:48Z

Resolves #3129

Related to open-telemetry/opentelemetry.io#8331

Signed-off-by: Pavol Loffay <[email protected]>

jpkrohling · 2025-11-17T11:25:22Z

projects/mcp-server.md

+
+### Industry outreach (Optional)
+
+Who (people, companies) in the industry should be aware of this effort? Was there an attempt to get them onboard? What did they say?


Can you share a bit about this? You listed existing projects, it would be nice to get in touch with them and share their feedback about this project

cc @mottibec, @adrielp

I have already talked to @austinlparker, any feedback is appreciated.

cc) @shiftyp

I will re-ping the other authors on slack.

Hey! What can I do to help!

@shiftyp would you like to join the SIG as a contributing member?

Sure! This is something I'm passionate about. Is the main channel for communication in the CNCF Slack? I have an account there

jpkrohling · 2025-11-17T11:25:51Z

projects/mcp-server.md

+#### Project Leads(s)
+
+* @pavolloffay (Red Hat)
+* Looking for other leads with experience in OpenTelemetry instrumentation and collector.


cc @niwoerner , would you be interested in collaborating?

I'd be interested in joining this project. I'll reach out to @pavolloffay on Slack

adrielp · 2025-11-17T14:58:08Z

Count me down for being a contributing member! Looking forward to this SIG. Here's a related thread in the #otel-semantic-conventions slack channel from a couple weeks ago that may be of interest to this SIG.

shiftyp · 2025-11-17T17:47:55Z

projects/mcp-server.md

+* [pavolloffay/opentelemetry-mcp-server](https://github.com/pavolloffay/opentelemetry-mcp-server): Focuses on collector configuration.
+* [austinlparker/otel-mcp](https://github.com/austinlparker/otel-mcp): Handles collector configuration and data profiling.
+* [mottibec/otelcol-mcp](https://github.com/mottibec/otelcol-mcp): Focuses on collector configuration.
+* [shiftyp/otel-mcp-server](https://github.com/shiftyp/otel-mcp-server): Provides data profiling, but requires OpenSearch.


Happy to answer any specific questions about this project!

Signed-off-by: Pavol Loffay <[email protected]>

pavolloffay · 2025-11-17T18:08:46Z

@niwoerner @shiftyp @adrielp I have added you to the proposal. Thanks!

shiftyp

Just some questions specifically related to the use case around data profiling, which I take to mean connecting telemetry itself to an agent flow, not just the configuration and troubleshooting use case for the OTEL stack itself.

shiftyp · 2025-11-17T19:21:04Z

projects/mcp-server.md

+- OpenTelemetry collector configuration
+
+Phase 2: Data profiling via collector (Months 1-2)
+- OpenTelemetry collector extension which provides API to query and profile the processed telemetry data


Just as context, this was the focus of my particular project. An MCP server that generated OpenSearch / ElasticSearch queries to feed relevant telemetry data into an agent (tested with Claude). I used it specifically with Claude Code to combine telemetry intelligence with code context to answer root cause question about incidents, analyze performance of code changes, ect. This is probably where I could contribute the most in terms of thought partnership, although my effort could be used towards various goals.

shiftyp · 2025-11-17T19:27:24Z

projects/mcp-server.md

+
+### Project Scope and Architecture
+
+The scope of this project is to create OpenTelemetry MCP server(s) to simplify deployment and management of the OpenTelemetry stack.


I find this statement slightly confusing, is this SIG simply to connect OTEL to an agent for configuration purposes? I think later the idea of collecting intelligence from telemetry is discussed (data profiling?), which should probably be indicated here if that's in scope.

I have slightly changed the paragraph

Great! This is better from my perspective

projects/mcp-server.md

Signed-off-by: Pavol Loffay <[email protected]>

niwoerner · 2025-11-19T19:29:09Z

projects/mcp-server.md

+
+The OpenTelemetry project consists of a large number of components, including collector, SDKs, and instrumentation libraries, which are often configured and managed separately. This distribution of components poses a major operational challenge which is universally recognized by the community [1], [2].
+
+Agentic workflows, powered by AI and language models, present a significant opportunity to simplify the deployment, configuration, and management of the OpenTelemetry stack. An agent could, for example, analyze telemetry data, suggest configuration changes to optimize performance, or automatically instrument new services.


I think the mentioned opportunities not only apply to agentic workflows, but to LLMs in general. While I like the mentioned examples, we could perhaps phrase it with a bit broader scope as it opens the door for additional use-cases.

Maybe something like the below suggestion?

Suggested change

Agentic workflows, powered by AI and language models, present a significant opportunity to simplify the deployment, configuration, and management of the OpenTelemetry stack. An agent could, for example, analyze telemetry data, suggest configuration changes to optimize performance, or automatically instrument new services.

Large language models (LLMs), present a significant opportunity to simplify the adoption, implementation, and management of the OpenTelemetry stack. An AI agent could, for example, analyze telemetry data, facilitate configuration changes, or assist and simplify the instrument process.

niwoerner · 2025-11-19T19:30:27Z

projects/mcp-server.md

+
+Agentic workflows, powered by AI and language models, present a significant opportunity to simplify the deployment, configuration, and management of the OpenTelemetry stack. An agent could, for example, analyze telemetry data, suggest configuration changes to optimize performance, or automatically instrument new services.
+
+However, to support this process, a standardized interface is required for an AI agent to interact with the OpenTelemetry ecosystem. The Model Context Protocol (MCP) provides an idiomatic approach for this interaction.


Suggested change

However, to support this process, a standardized interface is required for an AI agent to interact with the OpenTelemetry ecosystem. The Model Context Protocol (MCP) provides an idiomatic approach for this interaction.

However, to support this process, a standardized interface is required for LLMs to interact with the OpenTelemetry ecosystem. The Model Context Protocol (MCP) provides an idiomatic approach for this interaction.

niwoerner · 2025-11-19T19:32:33Z

projects/mcp-server.md

+* [mottibec/otelcol-mcp](https://github.com/mottibec/otelcol-mcp): Focuses on collector configuration.
+* [shiftyp/otel-mcp-server](https://github.com/shiftyp/otel-mcp-server): Provides data profiling, but requires OpenSearch.
+* [liatrio-labs/otel-instrumentation-mcp](https://github.com/liatrio-labs/otel-instrumentation-mcp): Manages instrumentation.
+* [traceloop/opentelemetry-mcp-server](https://github.com/traceloop/opentelemetry-mcp-server):  Provides data profiling by connecting to Jaeger, Tempo and Traceloop


Suggested change

* [traceloop/opentelemetry-mcp-server](https://github.com/traceloop/opentelemetry-mcp-server): Provides data profiling by connecting to Jaeger, Tempo and Traceloop

* [traceloop/opentelemetry-mcp-server](https://github.com/traceloop/opentelemetry-mcp-server): Provides data profiling by connecting to Jaeger, Tempo and Traceloop

niwoerner · 2025-11-19T19:47:11Z

projects/mcp-server.md

+* [liatrio-labs/otel-instrumentation-mcp](https://github.com/liatrio-labs/otel-instrumentation-mcp): Manages instrumentation.
+* [traceloop/opentelemetry-mcp-server](https://github.com/traceloop/opentelemetry-mcp-server):  Provides data profiling by connecting to Jaeger, Tempo and Traceloop
+
+Each of these servers uses a different approach, particularly for collector configuration and data profiling. This fragmentation creates confusion for users and hinders the development of a unified, agent-driven management plan for OpenTelemetry.


What is meant with agent-driven management plan ?

An additional downside of the fragmentation is that the setup becomes time consuming if every OTel MCP server has a different setup. The LLM response quality also likely decreases if more (potentially redundant) tool definitions are sent to the LLM which fills up the context window.

niwoerner · 2025-11-19T19:55:09Z

projects/mcp-server.md

+
+### Project Scope and Architecture
+
+The scope of this project is to create OpenTelemetry MCP server(s) to simplify deployment and day-2 operations.


As we'll likely would end up with different MCP servers for different use-cases, I'd love to establish an unified interface as part of this project.

Both, for a streamlined development of (multiple) servers AND a consistent way for the end-user to setup/configure the server(s)

niwoerner · 2025-11-19T20:17:35Z

projects/mcp-server.md

+The scope of this project is to create OpenTelemetry MCP server(s) to simplify deployment and day-2 operations.
+The MCP server should also provide data profiling/intelligence capabilities to support the day-2 operation use-cases.
+
+### Goals, objectives, and requirements


I like a lot of the goals we have here right now, but I believe it would be great if we could group/categorize them and specify what exactly we are trying to achieve.

Perhaps something like the below examples:

Collector:

deployment

configuration

...

Semantic Conventions:

context optimized querying of the official semconv registry

Weaver schema generation

...

Instrumentation:

Assisting with SDK initialization

Detecting broken traces

Identifying high cardinality issues

...

What do you think? I could work on rephrasing this section within the next days

Add Model Context Protocol (MCP) project proposal

8bb7343

Signed-off-by: Pavol Loffay <[email protected]>

pavolloffay requested review from a team, alolita, austinlparker, jpkrohling, maryliag, mtwo, mx-psi, svrnm, tedsuo and trask as code owners November 6, 2025 10:02

pavolloffay mentioned this pull request Nov 6, 2025

Onboard SIG: MCP Server #3129

Open

4 tasks

pavolloffay added 2 commits November 6, 2025 11:05

Add Model Context Protocol (MCP) project proposal

148122b

Signed-off-by: Pavol Loffay <[email protected]>

Fix CI

e3b9af2

Signed-off-by: Pavol Loffay <[email protected]>

jpkrohling reviewed Nov 17, 2025

View reviewed changes

svrnm mentioned this pull request Nov 17, 2025

Add OpenTelemetry MCP blog post open-telemetry/opentelemetry.io#8331

Open

shiftyp reviewed Nov 17, 2025

View reviewed changes

Update proposal

08b2ac3

Signed-off-by: Pavol Loffay <[email protected]>

shiftyp reviewed Nov 17, 2025

View reviewed changes

projects/mcp-server.md Show resolved Hide resolved

pavolloffay added 2 commits November 18, 2025 11:55

Added review comments

b5e2576

Signed-off-by: Pavol Loffay <[email protected]>

Added review comments

e06255b

Signed-off-by: Pavol Loffay <[email protected]>

niwoerner reviewed Nov 19, 2025

View reviewed changes


		### Industry outreach (Optional)

		Who (people, companies) in the industry should be aware of this effort? Was there an attempt to get them onboard? What did they say?


		### Project Scope and Architecture

		The scope of this project is to create OpenTelemetry MCP server(s) to simplify deployment and management of the OpenTelemetry stack.


		The OpenTelemetry project consists of a large number of components, including collector, SDKs, and instrumentation libraries, which are often configured and managed separately. This distribution of components poses a major operational challenge which is universally recognized by the community [1], [2].

		Agentic workflows, powered by AI and language models, present a significant opportunity to simplify the deployment, configuration, and management of the OpenTelemetry stack. An agent could, for example, analyze telemetry data, suggest configuration changes to optimize performance, or automatically instrument new services.

	Agentic workflows, powered by AI and language models, present a significant opportunity to simplify the deployment, configuration, and management of the OpenTelemetry stack. An agent could, for example, analyze telemetry data, suggest configuration changes to optimize performance, or automatically instrument new services.
	Large language models (LLMs), present a significant opportunity to simplify the adoption, implementation, and management of the OpenTelemetry stack. An AI agent could, for example, analyze telemetry data, facilitate configuration changes, or assist and simplify the instrument process.


		Agentic workflows, powered by AI and language models, present a significant opportunity to simplify the deployment, configuration, and management of the OpenTelemetry stack. An agent could, for example, analyze telemetry data, suggest configuration changes to optimize performance, or automatically instrument new services.

		However, to support this process, a standardized interface is required for an AI agent to interact with the OpenTelemetry ecosystem. The Model Context Protocol (MCP) provides an idiomatic approach for this interaction.

	* [traceloop/opentelemetry-mcp-server](https://github.com/traceloop/opentelemetry-mcp-server): Provides data profiling by connecting to Jaeger, Tempo and Traceloop
	* [traceloop/opentelemetry-mcp-server](https://github.com/traceloop/opentelemetry-mcp-server): Provides data profiling by connecting to Jaeger, Tempo and Traceloop

Add Model Context Protocol (MCP) project proposal #3128

Are you sure you want to change the base?

Add Model Context Protocol (MCP) project proposal #3128

Uh oh!

Conversation

pavolloffay commented Nov 6, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

adrielp commented Nov 17, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

pavolloffay commented Nov 17, 2025

Uh oh!

shiftyp left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

niwoerner Nov 19, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

niwoerner Nov 19, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

pavolloffay commented Nov 6, 2025 •

edited

Loading

adrielp commented Nov 17, 2025 •

edited

Loading

niwoerner Nov 19, 2025 •

edited

Loading

niwoerner Nov 19, 2025 •

edited

Loading