A skill for managing MCP tunnels through the local tunnel-client plugin. MCP, or Model Context Protocol, lets an agent connect to external tools and services.

Creating tunnel runtimes, connecting MCP servers over tunnels, listing aliases, checking runtime status, and stopping runtimes.

It provides commands and workflows for creating, connecting, inspecting, and stopping tunnel runtimes instead of managing those runtime processes manually.

tunnel-mcp — Skill for Claude Code and Codex by openai

About the project

openai/tunnel-client is a customer-run program that connects a private or local MCP server to ChatGPT, Codex, the Responses API, or AgentKit through an OpenAI-hosted tunnel. It is for operators who need these OpenAI products to reach an MCP server without exposing that server to the public internet.

openai/tunnel-client · 373 stars · on GitHub

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent

npx skills add openai/tunnel-client --skill tunnel-mcp

Clone the repo

git clone --depth 1 https://github.com/openai/tunnel-client

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/openai/tunnel-client/tunnel-mcp/github.svg)](https://agentmods.dev/skills/openai/tunnel-client/tunnel-mcp)

Your own site

<a href="https://agentmods.dev/skills/openai/tunnel-client/tunnel-mcp"><img src="https://agentmods.dev/badge/skills/openai/tunnel-client/tunnel-mcp/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/openai/tunnel-client/tunnel-mcp"><img src="https://agentmods.dev/badge/skills/openai/tunnel-client/tunnel-mcp.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 43 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 823 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits

NVIDIA SkillSpector pass 7 Sept 2026

How audits are shown

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00043	$0.00823
Opus 5	$0.00022	$0.00411
Sonnet 5	$0.00009	$0.00165
Haiku 4.5	$0.00004	$0.00082

Measured 10d ago against content hash 6b3da5547e6c, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

tunnel-mcp scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/tunnel-mcp/skills/tunnel-mcp/SKILL.md · 69 lines

How it starts

The opening of the file, as written. The whole thing — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Tunnel MCP

Use scripts/tunnel_mcp from this plugin when a user asks Codex to manage MCP tunnels through tunnel-client. The plugin entrypoint is a thin router onto the public native tunnel-client runtimes ... and tunnel-client admin-profiles ... command trees.

When the tunnel-mcp MCP app tools are available, use them first instead of manual shell routing:

install_or_select_tunnel_client
create_tunnel_runtime
connect_stdio_mcp
list_runtime_aliases
runtime_status
stop_runtime

The app tools are an operator surface over native tunnel-client; they orchestrate tunnel-client runtimes ..., normalize structured output, and keep tunnel protocol/runtime behavior in the Go binary.

Before acting, consult only the smallest relevant reference under references/:

references/binary.md: how to find or obtain a public-safe tunnel-client binary
references/setup-and-install.md: install, export, reset, binary-vs-bundle setup
references/profiles-state-and-keys.md: profiles, state dirs, admin/runtime key split
references/runtime-flows.md: create, connect, list, status, stop, rm, attach by tunnel id
references/troubleshooting.md: /healthz, /readyz, /ui, status, logs, stale aliases

Rules

Use tunnel-client admin tunnels for remote tunnel CRUD. Do not call raw tunnel-service HTTP endpoints from this plugin.
Route operational actions through tunnel-client runtimes ... and tunnel-client admin-profiles ....
Use scripts/tunnel_mcp self-check for plugin/binary/router compatibility; it must report secret reference presence without printing secret values.
Use native tunnel-client run --profile <name> only when the user intentionally wants a foreground daemon attached to the current terminal; do not translate profile files into flags in the plugin layer.
For a long-lived local runtime managed by Codex, use tunnel-client runtimes connect ...; do not use nohup or disown as the tunnel-client supervision path.
After runtimes connect, run tunnel-client runtimes status <alias> before reporting success. Only report success when status shows the managed runtime running with health reported; use --json when Codex needs explicit process_running, healthy, and ready fields.
Do not assume a source checkout, build system, helper, or tmux. The installed plugin must work with the selected tunnel-client binary alone.
Treat ambient PATH binary candidates as diagnostics unless selected through --tunnel-client-bin, TUNNEL_CLIENT_BIN, or .tunnel-client-bin.
Tunnel state, admin profiles, generated runtime profiles, stale-alias handling, cleanup classification, and local process management are owned by native tunnel-client; consult the relevant reference before explaining those details.
Keep admin and runtime credentials split: admin CRUD uses admin-profiles; runtime attach/connect uses --runtime-api-key env:NAME or file:/path. Do not pass literal keys.
Never write literal API keys, bearer tokens, cookies, or inline sk- style secret material into plugin state or generated configs.
Surface control_plane_poll_health separately from /healthz and /readyz; local readiness can be green while control-plane polling fails through a dead proxy.

Read the full file on GitHub · 69 lines

Files

What ships with it

5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

10d ago First seen · 69 lines · 43 tokens per session scan A 6b3da5547e6c

Subscribe to this mod's changes

tunnel-mcp is a skill published in the GitHub repository openai/tunnel-client (373 stars, last pushed today), licensed Apache-2.0. It adds 43 tokens to every session and 823 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Building agents DevOps and cloud agentic agentic-ai agentic-workflow chatgpt cli codex

skillpack-creator

Create a reusable SkillPack from a successful completed task. Use when the user wants to convert a one-off research, coding, analysis, or content workflow into a distributable local SkillPack with skillpack.json, local skills under skills/, starter prompts, start scripts, and an optional zip package.

CreminiAI/skillpack · 67 tokens

skill-creator

Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

CreminiAI/skillpack · 64 tokens

tunnel-mcp