Integration with MCP (Model Context Protocol) servers using OpenAI agents in Temporal workflows.

Adapted from OpenAI Agents SDK MCP examples

Before running these examples, be sure to review the prerequisites and background on the integration.

First, start the worker:

uv run openai_agents/mcp/run_file_system_worker.py

Run the workflow:

uv run openai_agents/mcp/run_file_system_workflow.py

This sample assumes that the worker and run_file_system_workflow.py are on the same machine.

First, start the worker:

uv run openai_agents/mcp/servers/tools_server.py --transport=streamable-http

Then start the worker:

uv run openai_agents/mcp/run_streamable_http_worker.py

Finally, run the workflow:

uv run openai_agents/mcp/run_streamable_http_workflow.py

First, start the MCP server:

uv run openai_agents/mcp/servers/tools_server.py --transport=sse

Then start the worker:

uv run openai_agents/mcp/run_sse_worker.py

Finally, run the workflow:

uv run openai_agents/mcp/run_sse_workflow.py

First, start the MCP server:

uv run openai_agents/mcp/servers/prompt_server.py

Then start the worker:

uv run openai_agents/mcp/run_prompt_server_worker.py

Finally, run the workflow:

uv run openai_agents/mcp/run_prompt_server_workflow.py

Demonstrates durable note-taking with the Memory MCP server: write seed notes, query by tags, synthesize a brief with citations, then update and delete notes.

Start the worker:

uv run openai_agents/mcp/run_memory_research_scratchpad_worker.py

Run the research scratchpad workflow:

uv run openai_agents/mcp/run_memory_research_scratchpad_workflow.py