AnyiWang

AnyiWang / OpenCovibe

Public

Local-first desktop app for AI coding agents (Claude Code, Codex). Built with Tauri v2 + Svelte 5.

ai-coding claude-code desktop-app developer-tools rust
54
7
100% credibility
Found Mar 03, 2026 at 54 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Svelte
AI Summary

OpenCovibe is a local-first desktop application that wraps AI coding CLIs like Claude Code with a native UI, providing features such as visual tool cards, run history and replay, multi-provider switching, agent management, and usage analytics while keeping all data stored locally.

Star Growth

See how this repo grew from 54 to 54 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is OpenCovibe?

OpenCovibe is a local-first desktop app built with Tauri v2 and Svelte 5 that wraps AI coding agents like Claude Code and Codex. It turns terminal-only CLIs into a full GUI experience with visual tool cards, syntax-highlighted diffs, session history replay, and multi-provider switching—all data stored locally with no cloud backend. Developers get persistent dashboards for runs, usage analytics, and inline permissions without losing CLI power.

Why is it gaining traction?

It stands out by layering desktop UX on raw CLI agents: browse past sessions, fork from any point, track token costs with heatmaps, and hot-swap 15+ providers like DeepSeek or Ollama mid-session. The setup wizard auto-detects CLIs and handles auth, making it dead simple to onboard. Local-first github fans appreciate the no-remote-data vibe in a sea of cloud-heavy AI tools.

Who should use this?

Claude Code CLI users frustrated by ephemeral terminal sessions and lacking visuals for tool outputs or diffs. Perfect for macOS devs running AI-assisted coding workflows who want run history, team dashboards, and plugin marketplaces without switching contexts. Suited for solo coders or small teams experimenting with multi-agent setups.

Verdict

Early project at 54 stars and 1.0% credibility score—docs are strong with pre-built macOS binaries, but Windows/Linux need more testing. Worth a spin for Tauri+Svelte enthusiasts seeking a local-first desktop front-end for Claude Code; contribute if you need cross-platform polish.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.