awesome-repositories.com
Blog
MCP
awesome-repositories.com

Discover the best open-source repositories with AI-powered search.

ExploreCurated searchesOpen-source alternativesSelf-hosted softwareBlogSitemap
ProjectMCP serverAboutHow we rankPress
LegalPrivacyTerms
© 2026 Bringes Technology SRL·VAT RO45896025·hello@awesome-repositories.com
omnirush-ai avatar

omnirush-ai/omnirush-gui

0
View on GitHub↗

Omnirush Gui

Omnirush GUI is a desktop application that runs an AI coding agent against files on a person's own machine, along with Git repositories and a web browser. Each chat can use its own model and reasoning effort level, chosen from the models available in the connected account. Local data stays local by default, with optional cloud use. Git and GitHub tasks such as cloning, branching, committing, pushing and opening pull requests can be run from chat, including parallel branches through separate worktrees.

The project puts heavy emphasis on controlling what the agent may do. Shell commands are classified by risk tier, with read-only commands running freely and riskier ones prompting for approval. Destructive actions such as force pushes, hard resets and recursive deletes always ask first. Users can switch between a guarded mode and a full-permissions mode, and organization-defined deny rules take precedence over either. Freshly opened projects are blocked from running their own plugins, tools and language servers by default, and an optional Linux sandbox keeps agent-driven builds and tests off the user's machine.

Around the desktop client sits an organization layer. A central control plane provisions inference, manages teams and access, and applies policies to every installation. Skills and plugins can be distributed through marketplaces so members receive only approved capabilities, and a single MCP endpoint exposes those assigned capabilities to compatible clients such as Codex, Claude Code and Cursor. Automated security review runs on pull requests and uncommitted local changes, withholding approval while blocking findings remain and scanning for confidential identity leaks.

The repository also covers its own delivery and testing. It builds installers for macOS (Apple Silicon and Intel), Windows and Linux, publishes tagged releases, and delivers checksum-verified in-app updates with rollback to earlier releases. End-to-end tests run in reproducible local or cloud sandboxes and capture screenshots and recordings as evidence of passing behavior.

AI search

Explore more awesome repositories

Describe what you need in plain English — the AI ranks thousands of curated open-source projects by relevance.

Start searching with AI
omnirush.ai
↗

Features

  • Desktop AI Agents - Desktop application running an AI coding agent against local files, Git repositories and a web browser, with per-chat model and effort selection.
  • Tool Execution Approvals - Asks for user confirmation before risky agent tool calls, always prompts on destructive actions, and blocks untrusted project code from running by default.
  • Local File Chat Agents - Runs an AI agent against files on the user's own machine, keeping data local by default with optional cloud use.
  • Agent Action Approval Policies - Controls what the agent may run by asking before risky commands, always prompting on destructive actions, and blocking untrusted project code by default.
  • Chat-Driven Git and GitHub Operations - Runs cloning, branching, worktrees, commits, pushes, and pull request creation or review with git and gh directly from the chat.
  • Team Management - Manages teams, access and desktop policies from one central control plane that applies to every desktop installation.
  • Command Risk Classification - Classifies every shell command segment as read-only, write, destructive, or interpreter, running safe ones freely and prompting before riskier ones.
  • Agent Tool Permissions - Asks permission before write or destructive agent actions and offers a full-permissions toggle that switches between prompting and automatically allowing every tool call.
  • Skill Installation - Installs reusable skills from a built-in library in settings so every available model sees the same installed skills.
  • Reasoning Effort Configurations - Lets each chat pick its own model from the connected account and set a reasoning effort level.
  • Per-Chat Model Selection - Lets each chat use its own model from the connected account and its own low, high, xhigh or max reasoning effort level.
  • Action Approval Gates - Always asks for approval before destructive actions such as force pushes, hard resets, branch deletions, and recursive deletes, even when nested in wrappers or subshells.
  • Skill Marketplaces - Distributes curated skills and plugins through organization marketplaces so members receive only approved capabilities.
  • Browser Automation Agents - Lets an AI agent drive a web browser through multi-step tasks on a person's behalf from the desktop workspace.
  • MCP Capability Management - Exposes assigned skills, plugins and connections through one MCP endpoint that clients like Codex, Claude Code and Cursor can call.
  • MCP Gateways - Offers a single MCP endpoint exposing assigned skills, plugins and connections so compatible clients can discover and call them.
  • Agent Sandboxes - Offers an optional Linux sandbox that keeps agent-driven builds, tests, and tool servers off the user's machine.
  • Agentic Plugin Marketplaces - Distributes curated skills and plugins through marketplaces so organization members receive only approved agent capabilities.
  • Automated Pull Request Reviewers - Automatically reviews pull requests and uncommitted diffs for security issues and identity leaks, blocking approval while findings remain.
  • Pull Request Review Interfaces - Runs automated security review on pull requests and local changes, withholding approval while blocking findings remain and scanning for confidential identity leaks.
  • Sandboxed Shell Executions - Optional Linux sandbox runs agent-driven builds, tests and tool servers off the user's machine, with risk-tiered approvals for shell commands.
  • Desktop App Packaging and Auto-Updates - Builds installers for macOS, Windows and Linux, publishes tagged releases, and delivers checksum-verified in-app updates with rollback to earlier versions.
  • Cross-Platform Installations - Produces installers and archives for macOS on Apple Silicon and Intel, Windows and Linux, each built independently within one release.
  • Tag-Triggered Release Automators - Triggers an automated workflow from a semantic-version tag to build installers per platform, generate checksums, and publish only after every build succeeds.
  • Approval Mode Switching - Lets users switch between guarded mode that prompts on risky tool calls and full-permissions mode, reloading the engine so new rules apply within seconds.
  • Automated Diff Security Reviews - Runs automated security review on every pull request and on uncommitted local changes, withholding approval while blocking findings remain.
  • Merge Approval Gating - Withholds pull request approval while blocking findings remain, treats lower-severity findings as advisory, and double-checks results with a strong model.
  • Central Access Policies - Applies organization-wide deny rules and command policies centrally so every desktop installation follows the same restrictions and approved capabilities.
  • Organization Policy Overrides - Applies organization deny rules, blocked commands, and approved browser origins on top of approval modes so a deny always wins, even in full-permissions mode.
  • Organization Command Deny Rules - Appends organization-defined deny and blocked-command rules after the defaults so an organizational restriction always overrides any local or remembered approval.
  • Untrusted Project Code Blocking - Blocks a freshly opened project from running its own plugins, custom tools, language servers and formatters by default, showing a notice of what was skipped.
  • Centralized Policy Administration - Central control plane that provisions inference, manages teams and access, and enforces approval and command policies on every desktop installation.
  • Agent Skill Extenders - Adds reusable skills and compatible plugins that give agents new abilities and task-specific workflows inside the desktop app.
4,876 stars·140 forks·TypeScript·9 views

Star history

Star history chart for omnirush-ai/omnirush-guiStar history chart for omnirush-ai/omnirush-gui

How this analysis was created: This summary and feature list are AI-generated from collected project material and can contain mistakes. Stars, license and language are imported from GitHub. Inclusion does not mean that we have tested or audited this project. Check the source documentation for any feature you depend on. Learn more on our About page.

Frequently asked questions

What does omnirush-ai/omnirush-gui do?

Omnirush GUI is a desktop application that runs an AI coding agent against files on a person's own machine, along with Git repositories and a web browser. Each chat can use its own model and reasoning effort level, chosen from the models available in the connected account. Local data stays local by default, with optional cloud use. Git and GitHub tasks such as cloning, branching, committing, pushing and opening pull requests can be run from chat, including parallel branches…

What are the main features of omnirush-ai/omnirush-gui?

The main features of omnirush-ai/omnirush-gui are: Desktop AI Agents, Tool Execution Approvals, Local File Chat Agents, Agent Action Approval Policies, Chat-Driven Git and GitHub Operations, Team Management, Command Risk Classification, Agent Tool Permissions.

Which projects share features with omnirush-ai/omnirush-gui?

Projects with overlapping indexed features include: totec448-spec/chat-on-steroids — Chat on Steroids is a desktop companion app that connects ChatGPT to the user's own machine through local Model… thunderbird/thunderbolt — Thunderbolt is an open-source, self-hostable AI chat client that runs on web, desktop (macOS, Linux, Windows) and… liuup/claude-code-analysis — This repository is a static analysis of the Claude Code terminal coding assistant, based on a reading of its… x-cmd/x-cmd — x-cmd is an AI agent orchestrator, cloud infrastructure CLI, and cross-platform package manager that provides an… openhands/openhands — OpenHands is an autonomous agent framework designed for software engineering workflows. It provides a modular platform… lintsinghua/claude-code-book — This repository is a long-form technical book about how a production AI coding agent runtime, often called an agent…

Projects sharing features with Omnirush Gui

These projects share indexed features with Omnirush Gui. Shared tags can include platform or build tooling; verify the primary use case before treating a result as a replacement.
  • totec448-spec/chat-on-steroidstotec448-spec avatar

    totec448-spec/chat-on-steroids

    5,144View on GitHub↗

    Chat on Steroids is a desktop companion app that connects ChatGPT to the user's own machine through local Model Context Protocol servers. It lets the chat read, search and patch files in a real project, run tests, keep terminals open, and observe or control the desktop, with a companion Chrome extension that drives real browser tabs in the background without changing the user's selected tab or foreground window. Access is split across separate Core, Desktop and Plugins connections, each authorized independently. What sets it apart is the permission layer around that access. File access is res

    TypeScriptautomationchatgptchrome-extension
    View on GitHub↗5,144
  • thunderbird/thunderboltthunderbird avatar

    thunderbird/thunderbolt

    4,819View on GitHub↗

    Thunderbolt is an open-source, self-hostable AI chat client that runs on web, desktop (macOS, Linux, Windows) and mobile (iOS, Android) from one shared web codebase inside lightweight native shells. It is local-first: each device keeps a local database as the source of truth, and data syncs to a central backend with optional end-to-end encryption, so the server stores only ciphertext for protected content. New devices are approved from trusted ones by fingerprint, recovery phrase or passkey, and revoked devices lose access. What sets it apart is its approach to model and agent choice and to p

    TypeScriptaiai-agentsllms
    View on GitHub↗4,819
  • liuup/claude-code-analysisliuup avatar

    liuup/claude-code-analysis

    4,069View on GitHub↗

    This repository is a static analysis of the Claude Code terminal coding assistant, based on a reading of its TypeScript source. It documents how the agent is built, from its startup path and a single shared execution core to the streaming tool-call loop, sub-agents, and terminal interface layers. It is a reference and study resource about the agent rather than the agent itself. The analysis centers on a few areas. One is the shared core that routes queries, sub-agents, tools, and permission checks through the same control flow. Another is memory: layered, file-based records that users can ins

    TypeScript
    View on GitHub↗4,069
  • x-cmd/x-cmdx-cmd avatar

    x-cmd/x-cmd

    4,739View on GitHub↗

    x-cmd is an AI agent orchestrator, cloud infrastructure CLI, and cross-platform package manager that provides an enhanced POSIX shell toolkit. It integrates large language models directly into the terminal for chatting, code generation, and the execution of agentic workflows, while offering a framework for building interactive terminal user interface components. The project distinguishes itself by deploying containerized AI agents within isolated sandboxes, provisioning them with specialized skills and headless browser automation capabilities. It further streamlines development through a unif

    Awkagentaibash
    View on GitHub↗4,739
Compare all 30 related projects→