A native DSH plugin that bridges in-browser ChatGPT to your local DeepSeek Harness project: direct file/command access plus native DSH Session handoff for long-running work.
DSH integration
Native runtime
Author-claimed
Safety audit
Unaudited
Last verified
2026-09-04
License
MIT
01What can it help you accomplish?
Let in-browser ChatGPT read, edit and run a real local codebase instead of copying context in and out of the chat.
File reads/edits, command and engineering-tool runs, plus Diagnostics, Git status, and build/test/exec results — all against the actual project.
Developers who use ChatGPT in the browser and want it to operate directly on their local DeepSeek Harness projects.
Hand a heavy or long-running task from ChatGPT over to a native DSH Session to keep executing.
DSH continues the edits/build/test loop; ChatGPT can later re-inspect the code, diffs, test results and other execution results.
Users whose tasks need sustained multi-step edits, builds or tests beyond a single ChatGPT turn.
02How to install into DeepSeek Harness
Installation steps
- 01
Install the plugin with the dsh CLI: `dsh plugin --profile web add dsh-with-chatgpt`
$ dsh plugin --profile web add dsh-with-chatgpt
- 02
Start DSH in web mode: `dsh web`
$ dsh web
Verify the integration
Not specified by the author
03DSH integration and capability boundaries
Native DSH plugin: in-browser ChatGPT drives your local project via Agent Helm and can spawn native DSH Sessions for long-running work.
Agent Helm — local project access
an in-browser ChatGPT session connected to a local project→project-structure understanding, file find/read/edit, command runs, Diagnostics and Git status, build/test/exec checks
edits files in the connected local projectruns commands and engineering tools on your machineNative DSH Session handoff
a task ChatGPT already understands from the project context→a native DSH Session that keeps executing; ChatGPT can re-check code, diffs and test results
In-DSH ChatGPT activity view
ChatGPT's local work inside a DSH→view of the linked conversation, project/worktree, local operations, status, activity and linked DSH Session
04Who is it for? When not to use it?
Good for
- Developers who use ChatGPT in the browser and want it to operate directly on their local DeepSeek Harness projects.
- Users whose tasks need sustained multi-step edits, builds or tests beyond a single ChatGPT turn.
Not for
- The plugin lets ChatGPT edit files and run commands on your machine; review operations before approving and scope access to the intended project/worktree.
05Compatibility, maintenance and safety notes
- On first run the plugin checks the environment and walks you through required installation, connection and authorization via the Setup Guide before ChatGPT can access the local project.
- The plugin lets ChatGPT edit files and run commands on your machine; review operations before approving and scope access to the intended project/worktree.
MIT · actively maintained (latest release v0.1.4, 2026-09-03)
06Frequently asked questions
What is DSH with ChatGPT?
It is a native DeepSeek Harness (DSH) plugin. You keep working in ChatGPT in the browser, but it can now understand your real local project, edit files, run commands and check results, and hand longer tasks to a native DSH Session.
How do I install it?
Install the plugin with the dsh CLI: `dsh plugin --profile web add dsh-with-chatgpt`, then start DSH with `dsh web`. On first run, a Setup Guide checks your environment and walks you through the needed installation, connection and authorization.
Does ChatGPT need my code pasted into the chat?
No. Through Agent Helm, in-browser ChatGPT works on the real project on your machine directly — understanding structure, finding and reading files, editing them, running commands, and viewing Diagnostics, Git status and build/test results — so you don't keep copying code, errors and context into the conversation.
When does it create a DSH Session?
When a task needs heavy edits, builds, tests or sustained execution, ChatGPT opens a native DSH Session that continues the work; afterward ChatGPT can re-inspect the code, diffs and test results in the real project before deciding the next step.
Can I see what ChatGPT did locally?
Yes. DSH with ChatGPT adds a lightweight entry inside DSH showing the linked ChatGPT conversation, the project/worktree used, the local operations performed, current status and recent activity, and whether a DSH Session is attached.
07Related DSH workflows
ruflo
by ruvnet
🌊 The original agent meta-harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, RAG integration, and native Claude Code / Codex / Hermes and many more Integrated
reactive-resume
by amruthpillai
A one-of-a-kind resume builder that keeps your privacy in mind. Completely secure, customizable, portable, open-source and free forever. Try it out today!
everos
by evermind-ai
One portable memory layer for every AI agent: local-first, Markdown-native, user-owned, and self-evolving across apps, tools, and workflows.
yao
by yaoapp
✨ All your agents and workspaces in one place, on every device you own. Track tasks on a board, accessible from desktop, mobile, browser, or API. Self-hosted.
08Data and sources
**DSH with ChatGPT** 是一个 DeepSeek Harness（DSH）Plugin。
当任务需要大量修改、构建、测试或者持续执行时，ChatGPT 可以在已经理解项目和任务的基础上，直接创建原生 DSH Session。
This page is generated from the project’s public documentation, repository metadata and a structured parse of DSH Plugins; last verified on 2026-09-04. Found an error? Submit a correction.
Best DeepSeek Harness Plugins
Twelve plugins worth installing first — picked from the whole catalog, across every category.