engineering / agent

Terax

A terminal-first development environment with a code editor, file explorer, source control and an AI agent panel that uses your own API keys or local models.

Repository owner crynta

Not rated yet
0 comments

Like 0 Save

Published 11 Oct 2026Listing updated 12 Oct 2026

At a glance

Purpose: A terminal-first development environment with a code editor, file explorer, source control and an AI agent panel that uses your own API keys or local models.
Runs in: Local agent with shell
Operating systems: macOS, Linux, Windows
Needs: Runtime: Not stated; Needs internet: No; Needs accounts or API keys beyond your agent: No
Files and data: Reads files: Yes; Writes files: Yes; Sends data to other services: Not stated
Tool type: Agent
Delivery: From GitHub
Version: Latest on the main branch
Access: Code on GitHub
Compatibility: According to the README, Terax runs on macOS, Linux and Windows. The README lists zsh, bash, pwsh, fish and cmd for the terminal backend. The README describes Windows shell detection in the order PowerShell 7+, Windows PowerShell 5.1, then cmd.exe, and per-tab workspaces for Local or any installed WSL distro. The README names an Arch AUR package, a Nix flake, an AppImage, and .deb and .rpm packages for Linux. The README names these AI providers: OpenAI, Anthropic, Google (Gemini), Groq, xAI (Grok), Cerebras, OpenRouter, DeepSeek and Mistral, plus any OpenAI-compatible endpoint. It names LM Studio, MLX and Ollama for local models.
External costs: According to the README, the AI panel uses your own API keys for providers such as OpenAI, Anthropic, Google (Gemini), Groq, xAI (Grok), Cerebras, OpenRouter, DeepSeek and Mistral, or a local model through LM Studio, MLX or Ollama. The README also says there is no account. The repository is Apache-2.0-licensed and states no pricing. The model provider you connect may charge for usage.
Permissions: The README describes desktop installers and auto-updates from GitHub Releases, Linux package-manager installation and pnpm dependency installation for source builds. The README says users supply provider API keys, which are stored in the OS keychain through keyring, or configure a local model endpoint. The README describes file-reading and file-changing tools, approval-gated Bash, background processes and Claude Code orchestration. The README says the app runs native shell sessions and provides staging, commits and pushes through source control. The README describes file-change updates, voice input and external URLs opened in a native child webview. The repository file list includes a scripts/ folder; the README does not describe its contents.
Usage terms: Apache License 2.0, as stated in the repository. See its LICENSE file for the exact terms.
Interface: desktop

Code on GitHub

crynta/terax-ai

Apache-2.0

TypeScript9,310 stars1,024 forksLast pushed 12 Oct 2026 UTC

Lightweight (7MB) Terminal-first AI-native dev workspace

Topics: agents ai code-editor linux macos reactjs rust tauri

Public GitHub data fetched 13 Oct 2026 UTC · 0 days old

View on GitHub ↗

What it does

The README describes Terax as a terminal-first development environment built on Tauri 2, Rust and React 19. The README lists a terminal, code editor, file explorer, source control with a Git graph and a web-preview pane.

The README describes an agentic AI side-panel using the user's API keys or local models. The README names OpenAI, Anthropic, Gemini, Groq, xAI, Cerebras, OpenRouter, DeepSeek and Mistral, plus OpenAI-compatible endpoints, and lists LM Studio, MLX and Ollama for local use.

The README lists plans, subagents, project memory in TERAX.md, file tools, approval-gated Bash, background processes and custom agents. The README also describes spawning Claude Code in a terminal and sending follow-up work through approval-gated tools.

Who it is for

Developers who want a terminal, editor and AI agent panel in one app and who bring their own model provider or a local model.

Example tasks

Split a terminal tab into horizontal and vertical panels.
Accept or reject AI edit diffs in the code editor hunk by hunk.
Stage and unstage hunks, commit and push from the source control panel.
Point the AI settings at a local LM Studio, MLX or Ollama endpoint.
Open a local development server in the web preview tab.

Getting started

The README points to installers on GitHub Releases and says Terax updates from there. The README gives yay -S terax-bin for Arch and describes Nix installation through its flake. The README says to open Settings -> AI, choose a provider and enter an API key, or configure an LM Studio, MLX or Ollama endpoint. The README lists stable Rust, Node 22+, pnpm and platform-specific Tauri prerequisites for source builds, followed by pnpm install and pnpm tauri dev or pnpm tauri build. The README has the exact commands.

Limitations

The README lists macOS, Linux and Windows. The README describes Windows shell detection and per-tab environments for Local or an installed WSL distribution. The README lists Linux AUR, Nix, AppImage, .deb and .rpm routes. The README says AppImage needs FUSE, provides --appimage-extract-and-run as a fallback and suggests WEBKIT_DISABLE_DMABUF_RENDERER=1 for Wayland rendering glitches. The README says the AI panel uses the user's API keys or local models. The README lists stable Rust, Node 22+, pnpm and platform-specific Tauri prerequisites for source builds. The repository can change at any time; check its README before you install.

Data practices

The README says Terax has no telemetry and requires no account. The README says provider API keys are stored in the OS keychain through keyring. Data destinations depend on the provider or local endpoint you configure. The README says files and selections can be attached to the AI side-panel and that Terax auto-updates from GitHub Releases.