openai/tunnel-client — Customer-run client for Secure MCP Tunnel: connect private or localhost MCP servers to ChatGPT, Code…

Description and use

**Project Summary:**
The openai/tunnel-client is an open-source project that enables secure communication between private or localhost servers and AI services like ChatGPT, Codex, and AgentKit, without exposing them to the public internet. This allows users to create repeatable agent workflows by coordinating multi-agent processes and integrating tools into pipelines. The project is production-ready, with a strong adoption rate, recent updates, and a well-maintained Go codebase.

**Value Proposition:**
The openai/tunnel-client provides a secure and standardized way to connect private servers to AI services, enabling users to create complex workflows and integrate tools into their pipelines. This can help organizations automate repetitive tasks, improve agent memory, and standardize their workflows.

**Practical Adoption Path:**
To adopt the openai/tunnel-client, users can follow these steps:

1. Install the Go package and set up the client.
2. Configure the client to connect to their private server and AI services.
3. Integrate tools and pipelines into their workflows.
4. Test and refine their workflows to achieve the desired outcomes.

**Production Readiness:**
The openai/tunnel-client is production-ready, with a high score of 85/100. The project has a strong adoption rate, with 220 GitHub

Problem

openai/tunnel-client helps turn isolated prompts and tools into repeatable agent workflows.

Outcome

High for an OSS candidate: recent activity, adoption, and ecosystem signals are strong enough for a serious pilot.

Workflow

• coordinate multi-agent workflows
• add tool-use pipelines
• standardize agent memory

Evidence

Customer-run client for Secure MCP Tunnel: connect private or localhost MCP servers to ChatGPT, Codex, the Responses API, and AgentKit without exposing them to the public internet.

Source