# AI use cases for Openchatx MCP (XiaoPuOuO/openchatx-mcp) Repo: [XiaoPuOuO/openchatx-mcp](https://github.com/XiaoPuOuO/openchatx-mcp) | Use cases: 4 | Page: https://dedupedai.com/use-cases/xiaopuouo-openchatx-mcp/ OpenChatX is a local agent runtime that gives ChatGPT tools to edit files, run shell commands, control your computer, and use MCP servers through one official MCP connection. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Let ChatGPT edit code and run shell commands locally: https://dedupedai.com/use-cases/xiaopuouo-openchatx-mcp/let-chatgpt-edit-code-on-your-machine/ - Continue long ChatGPT tasks in a fresh conversation: https://dedupedai.com/use-cases/xiaopuouo-openchatx-mcp/continue-long-tasks-in-new-chats/ - Enforce coding rules and reusable workflows for ChatGPT: https://dedupedai.com/use-cases/xiaopuouo-openchatx-mcp/apply-team-rules-and-skills-to-agent-work/ - Route existing MCP servers through one ChatGPT connection: https://dedupedai.com/use-cases/xiaopuouo-openchatx-mcp/route-one-chatgpt-connection-to-existing-mcp-servers/ ## How to let ChatGPT edit code and run shell commands locally with Openchatx MCP Repo: [XiaoPuOuO/openchatx-mcp](https://github.com/XiaoPuOuO/openchatx-mcp) | Difficulty: Intermediate | Time: 1 hour | Basis: From the README Give ChatGPT coding-grade file and shell tools on your own computer through one MCP connection. **Who it is for:** A developer who wants ChatGPT to read, change, and test a local project without copying code back and forth by hand. **The problem:** ChatGPT in a chat window cannot touch the files and commands on your machine, so every change needs manual copying. **The outcome:** ChatGPT calls OpenChatX file and shell tools against a bound project folder and reports the results in the same chat. ### Steps 1. Download the matching installer from https://github.com/XiaoPuOuO/openchatx-mcp/releases/latest and install OpenChatX Desktop. 2. In OpenAI Platform, create a tunnel on the Tunnels page and copy the tunnel_... ID. 3. Create a Runtime API key on the API Keys page and store it safely. 4. Open OpenChatX Desktop, go to More, choose Connect Tunnel, and enter the Tunnel ID and control-plane API key. 5. In ChatGPT, finish the MCP app by selecting the openchatx tunnel, setting authentication to No authentication, and clicking Create. ### Prompt to paste into your agent ```text Using the OpenChatX MCP connection in the openchatx-mcp repo, open the project at as a Projects workspace and list its files, then run its test command through the shell tool. Report which files changed and the test result. Verify that the shell output matches what you see in the project folder. ``` ### Caveat OpenChatX runs with your local user permissions, so a trusted ChatGPT caller can change files and run commands on the machine. Risk, credentials: The Runtime API key controls the Secure MCP Tunnel, so store it only in the OS credential store and never paste it into chat or commit it to Git. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/xiaopuouo-openchatx-mcp/let-chatgpt-edit-code-on-your-machine/ ## How to continue long ChatGPT tasks in a fresh conversation with Openchatx MCP Repo: [XiaoPuOuO/openchatx-mcp](https://github.com/XiaoPuOuO/openchatx-mcp) | Difficulty: Intermediate | Time: 1 hour | Basis: From the README Hand off a long-running task to a new chat using a stored summary and a UUID. **Who it is for:** A person working on a multi-day task in ChatGPT who has hit the practical context limit of one conversation. **The problem:** When a chat gets too long, the work state is lost unless you rebuild the context by hand. **The outcome:** A new ChatGPT conversation receives the stored summary and recent context from the old one, and the agent resumes the task. ### Steps 1. Complete the Connect OpenChatX to ChatGPT setup so the MCP connection works. 2. In the old chat, ask the Agent to call summarize and save the returned UUID. 3. Start a fresh ChatGPT conversation. 4. Give the Agent the UUID and ask it to call summarize with that UUID. 5. Check the Dashboard Summaries view to confirm the temporary summary exists before you delete it. ### Prompt to paste into your agent ```text Using the openchatx-mcp repo, call summarize at the end of the current task and return the UUID. In a new chat, call summarize with the UUID and continue the task at . Verify that the reply names the last completed step from the stored summary. ``` ### Caveat The handoff is deleted automatically after it is used, so save any important detail elsewhere before you start the new chat. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/xiaopuouo-openchatx-mcp/continue-long-tasks-in-new-chats/ ## How to enforce coding rules and reusable workflows for ChatGPT with Openchatx MCP Repo: [XiaoPuOuO/openchatx-mcp](https://github.com/XiaoPuOuO/openchatx-mcp) | Difficulty: Intermediate | Time: 1 hour | Basis: From the README Store coding conventions as Rules and step-by-step workflows as Skills the Agent loads on demand. **Who it is for:** A team lead who wants every ChatGPT coding session to follow the same React or project conventions. **The problem:** Conventions in one person's head are not applied consistently when different people or chats ask ChatGPT for code. **The outcome:** The Agent loads matching Rules when it touches relevant files and loads a Skill's full instructions when the task calls for it. ### Steps 1. Open the Dashboard and go to the Toolbox that will own the rules. 2. Create a .mdc Rule under the Toolbox rules folder with a description, file globs, and alwaysApply set to false. 3. Choose the Rule mode: Always, Auto Attached, Agent Requested, or Manual. 4. Create a Skill with a SKILL.md file that has name, description, and Markdown instructions. 5. Ask the Agent to run skill_search and then skill_search with action=load for the exact Skill name. ### Prompt to paste into your agent ```text Using the openchatx-mcp repo, create a Rule in the Toolbox that applies to files matching , and create a Skill named . Then edit a file matching and report which Rule was applied. Verify that the Rule text appears in the Agent's working context before it edits the file. ``` ### Caveat Rules are stored as files inside the Toolbox, so review them before sharing a Toolbox with others. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/xiaopuouo-openchatx-mcp/apply-team-rules-and-skills-to-agent-work/ ## How to route existing MCP servers through one ChatGPT connection with Openchatx MCP Repo: [XiaoPuOuO/openchatx-mcp](https://github.com/XiaoPuOuO/openchatx-mcp) | Difficulty: Advanced | Time: 1 hour 30 minutes | Basis: From the README Add your MCP servers behind the single OpenChatX connection and sign in where required. **Who it is for:** A person who already runs several MCP servers and does not want to add a separate connection in ChatGPT for each one. **The problem:** Each MCP server normally needs its own setup in ChatGPT, and remote servers often need sign-in. **The outcome:** ChatGPT reaches the added MCP servers through OpenChatX, and the Dashboard shows which services need sign-in. ### Steps 1. Open the Dashboard and go to the MCP Servers area. 2. Add a local or remote MCP server, or ask ChatGPT to add it to OpenChatX. 3. If the server requires sign-in, complete the OAuth authorization prompt shown in the MCP management UI. 4. Check the Dashboard status to confirm the server is connected rather than failing at runtime. 5. Run npm run print-url in a source install to confirm the local MCP and tunnel URLs. ### Prompt to paste into your agent ```text Using the openchatx-mcp repo, add the MCP server at through OpenChatX, complete any OAuth prompt, and call one of its tools. Report the server status shown in the Dashboard and the tool result. Verify that the status is not an authentication-required state. ``` ### Caveat The README says OpenChatX runs with your local user permissions, so every MCP server you add is reachable by a trusted ChatGPT caller. Risk, credentials: OAuth credentials for connected MCP servers are stored in the OS secure store, so review which servers you authorize before you allow ChatGPT to use them. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/xiaopuouo-openchatx-mcp/route-one-chatgpt-connection-to-existing-mcp-servers/