# Calling workflows from MCP clients and getting real results back

**URL:** <https://community.solace.com/t/calling-workflows-from-mcp-clients-and-getting-real-results-back/4946>\
**Category:** Solace Agent Mesh\
**Created:** [November 11, 2026, 4:00am UTC](https://community.solace.com/t/calling-workflows-from-mcp-clients-and-getting-real-results-back/4946 "2026-11-11T04:00:54Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Tamimi](https://yyz1.discourse-cdn.com/solace/user_avatar/community.solace.com/tamimi/32/2190_2.png) [@Tamimi](https://community.solace.com/u/Tamimi)\
**Post date:** [November 11, 2026, 4:00am UTC](https://community.solace.com/t/calling-workflows-from-mcp-clients-and-getting-real-results-back/4946/1 "2026-11-11T04:00:54Z")

</div>

Hi folks,

The MCP entrypoint in Agent Mesh publishes your agents and workflows as tools that any MCP client can discover and call. Previously, when a client called a workflow, all it got back was one line saying the workflow completed successfully. Not much use if you wanted the actual answer.

Now the client gets the workflow’s output mapping back, as JSON text and as structuredContent, so it can read the fields directly instead of parsing text. Failures are handled properly too. A failed, cancelled or rejected run comes back as an MCP error result with isError set, and the text says which node failed and why. Input that doesn’t match the workflow’s input schema is rejected with a list of the required fields.

I tried it with a freight quote workflow: a tool node that prices the lane with no agent and no LLM, then a switch that sends high value quotes to a different agent for an approval note. I called it from the MCP Inspector and a small Python client, looked at the raw responses and the run trace in Activities, and then asked for an unsupported shipping mode to show the error path:

[](https://www.youtube.com/watch?v=gm-Y5_cWW8o)

Which MCP clients are you connecting to Agent Mesh, and what workflows would you expose to them? Share it in the thread.

If you want to give it a go, [download Solace Agent Mesh](https://solace.com/products/agent-mesh/). You’ll find the GitHub repo here:

> **[GitHub - SolaceProducts/solace-agent-mesh: Platform for building and running AI agents across...](https://github.com/SolaceProducts/solace-agent-mesh)**
>
> Platform for building and running AI agents across the full lifecycle — build, test, deploy, monitor, improve. Hosts your agents, runs their tools, and coordinates them over an event mesh, with sessions, artifacts, auth, evaluators, and observability built in.