voyager-crew/voyager

Enhancement suite for Gemini, AI Studio, Claude, ChatGPT & DeepSeek — plus a prompt manager for any website, DeepSeek Harness included. / 面向 Gemini、AI Studio、Claude、ChatGPT 与 DeepSeek 的增强套件；其中的提示词管理器可用于任意网站，如 DeepSeek Harness。

What it solves

Voyager is a browser extension designed to bring structure and organization to AI chatbot interfaces. It addresses the lack of native organization tools in web UIs for Gemini, Claude, ChatGPT, and DeepSeek, allowing users to manage large volumes of conversations and prompts more effectively.

How it works

It functions as a companion layer over existing AI web interfaces. It provides a set of productivity tools—such as folder hierarchies, prompt libraries, and navigation timelines—that integrate directly into the chatbot's UI. The extension also supports cloud synchronization via Google Drive or iCloud to keep data consistent across devices.

Who it’s for

It is built for researchers, developers, and power users who juggle dozens of AI threads, save frequent code snippets, and require a more organized way to interact with multiple LLM providers.

Highlights

  • Folder Organization: Create two-level folder hierarchies with drag-and-drop support for chats.
  • Prompt Vault: Save and reuse prompts across different AI sites and even local development environments.
  • Timeline Navigation: Visual nodes to jump between messages and manage conversation branches.
  • Cross-Platform Support: Works across Gemini, Claude, ChatGPT, and DeepSeek.
  • Chat Export: Export conversations to JSON, Markdown, or PDF.
  • Visual Customization: Add seasonal ambience effects like snow, rain, or sakura petals to the interface.

Related

  • Project
    datawhalechina/easy-vibe

    A comprehensive, multilingual learning map and tutorial series that teaches users how to build full-stack AI applications through AI-assisted "vibe coding."

  • Project
    kvcache-ai/ktransformers

    A flexible framework for efficient LLM inference and fine-tuning using CPU-GPU heterogeneous computing, specifically optimized for large MoE models on consumer hardware.

  • Project
    facebook/prophet

    An automatic forecasting procedure for time series data that uses an additive model to handle strong seasonality, holiday effects, and trend shifts.

  • Project
    openai/gpt-oss

    OpenAI’s gpt‑oss repo delivers reference code and tooling for the open‑weight 120 B and 20 B LLMs, including a special Harmony chat format, MXFP4 quantization, and back‑ends for PyTorch, Triton, Metal, vLLM, Ollama, and more. Licensed Apache‑2.0, it’s aimed at researchers and developers who want to run, fine‑tune, or build tool‑augmented agents with these models.