← DeepSeek Harness Plugins
DeepSeek Harness PluginManifest Valid

dsh-llm-chatgpt-web

ChatGPT Web LLM provider for DeepSeek Harness via browser automation and optional native MCP tools.

VisionUI & ProductivityTerminal & TUIDeveloper ToolsBrowser & WebSecurity & PolicySkills & WorkflowsRemote Execution
0GitHub Stars0ForksUpdated2026-09-16

Install

$ dsh plugin --profile web add github:twilightt1/dsh-llm-chatgpt-web

Plugin Overview & Capabilities

AI-assisted organization based on the public repository snapshot. The content must be grounded in source evidence and does not replace compatibility or security verification.

source-grounded
dsh-llm-chatgpt-web is a DeepSeek Harness provider plugin that enables using ChatGPT Web without an API key. It manages a headless-hidden or headed Chromium daemon to drive Temporary Chat pages for text generation, extracting streaming Markdown directly from the DOM. Additionally, it offers an opt-in native MCP transport on Unix for structured tool calls via OpenAI Secure MCP Tunnels and local broker communication.

Key Capabilities

  • Drives ChatGPT Web through a dedicated background Chromium daemon without requiring an API key
  • Provides text streaming using stateless Temporary Chat pages with DOM Markdown extraction
  • Supports multiple reasoning effort profiles (e.g., luna, think, light, medium, high, extra-high, pro)
  • Offers an opt-in Unix native MCP tool transport with local broker socket and stdio MCP façade
  • Persists session and browser storage state across turns

Useful For

  • Using personal ChatGPT Web Plus/Pro accounts as LLM backends for DeepSeek Harness agents
  • Running local agent workflows without incurring API credit costs
  • Experimenting with native MCP tool execution through ChatGPT Personalized connectors

Who It Fits

  • DeepSeek Harness developers seeking an alternative web-based LLM provider
  • Users with ChatGPT subscriptions wanting to utilize their web quota inside DSH
  • Developers testing experimental MCP connector integrations

Documented Limitations

  • Requires a local Google Chrome / Chromium installation and a one-time manual sign-in
  • Native MCP tool continuation is experimental and restricted to Unix platforms
  • Image input, custom temperature, and explicit stop sequences are unsupported
  • Reasoning thoughts are not separated from output in stream tokens
  • Susceptible to breaking changes if ChatGPT updates its web UI DOM structure

DSH Compatibility

Version-specific runtime evidence collected by DSH Plugin. A missing result means we have not tested that combination yet.

Not tested yetNo runtime compatibility tests have been published yet.

Security Signals

Objective signals discovered from package metadata and source inspection. These are not a guarantee that a plugin is safe.

Dependency Counts

package.json declares 5 runtime, 8 development, 3 peer, and 0 optional dependencies.

info
Dsh Bundle Declared

package.json declares DSH bundle metadata.

info
License Declared

GitHub reports the repository license as NOASSERTION.

info
Package Manifest Available

A root package.json was captured and can be inspected by the registry.

info
Source Available

Public GitHub source metadata is available for this registry snapshot.

info

Source & Registry Notes

Public provenance, Registry classification, and the latest source check for this entry, kept separate from runtime verification.

Source repository
twilightt1/dsh-llm-chatgpt-web
Registry source
GitHub · dsh-plugin topic
Registry classification
Plugin
Source checked
e752186 · 2026-09-18

This project is independently indexed from public source information. DSH Plugin is not affiliated with DeepSeek or the plugin author. Always check the author repository before installation.

Repository Activity

GitHub stars
0GitHub stars
Forks
0
Open issues
1
Last commit
2026-09-16
Last release
2026-09-10
For maintainers

Maintaining this plugin?

This listing is generated from public repository data. If you maintain this project, you can review the information and share this listing with your users if you find it useful.

Add to README
Listed on DSHPlugin.app