zhu1090093659/deepseek-pp

DeepSeek Web browser extension: AI agent workspace with MCP tools, memory, Skills, automation, web search, and conversation export.

What it solves

DeepSeek++ transforms the DeepSeek web interface into a comprehensive AI agent workspace. It addresses the limitations of a standard chat interface by adding long-term memory, tool integration (via MCP), browser control, and automation, allowing users to perform complex, multi-step tasks directly within their browser workflow.

How it works

It operates as an open-source browser extension for Chrome, Edge, and Firefox. It enhances the DeepSeek web page by injecting a sidebar for conversations, adding right-click context menus for text processing, and integrating with Model Context Protocol (MCP) servers to execute local or remote tools. It can also install native hosts to enable multimodal analysis (images/video) and shell execution for local file manipulation.

Who it’s for

  • Users who want to turn DeepSeek into a fully-fledged AI agent with tool-calling and automation capabilities.
  • Professionals needing to manage project-specific contexts, memories, and reusable "Skills" (task templates).
  • Power users who want to export their AI conversations into various formats (HTML, Markdown, PDF) for archiving.
  • Developers who want to integrate local scripts and tools into their DeepSeek chat experience via MCP.

Highlights

  • Agentic Workflow: Supports continuous task execution where the model decides the next step based on tool results.
  • MCP Integration: Connects to remote or local MCP services for extended tool capabilities.
  • Long-term Memory: Automatically saves and injects user preferences and project backgrounds into conversations.
  • Browser Control: Allows the AI to navigate, click, and input text in selected browser tabs.
  • Skill System: Provides a library of built-in and custom task templates (Skills) that can be imported from GitHub or local folders.
  • Automation: Supports scheduled tasks using cron or RRULE expressions to trigger DeepSeek prompts automatically.
  • Multimodal Support: Enables image and video analysis by integrating with OpenAI or Gemini via native hosts.

Related

  • Project
    EdgeTypE/better-deepseek

    A browser extension that enhances DeepSeek with persistent memory, file generation tools, and codebase management capabilities for complex projects.

  • Project
    DeepLifeStudio/DeepSeekAI

    A browser extension that integrates a private DeepSeek-powered AI co-pilot into any webpage, allowing users to interact with AI via text selection and floating chat windows.

  • Project
    eyalzh/browser-control-mcp

    An MCP server and Firefox extension that lets AI assistants securely manage tabs, search history, and read webpage content with explicit user consent.

  • Project
    AgentDeskAI/browser-tools-mcp

    An MCP server and Chrome extension that gives AI agents access to a user's active browser session for real-time debugging, network inspection, and Lighthouse audits.

  • Project
    lessweb/deepcode-cli

    Deep Code CLI is a Node‑based terminal assistant tuned for DeepSeek‑v4 models. Install via `npm i -g @vegamo/deepcode-cli`, configure it with `~/.deepcode/settings.json`, and use slash commands (e.g., `/new`, `/model`, `/skills`) to chat, run custom agent skills, handle images, connect to external services via MCP, and control permissions. It shares settings with a VS Code plugin and supports notifications, web‑search scripts, and any OpenAI‑compatible endpoint.