Skip to content
Alternatives
Software like VibeBox
What else does this job. Matched on what each project does, not on who links to whom.
Closest first
- vibe-collabgithub.comAI-powered project collaboration framework with three-mode CLI (IDE/interactive/agent)
- vibe-researchpypi.orgvibe-research is an open-source CLI tool that enables autonomous, fully-cited multi-agent research workflows directly in the terminal. It is designed for researchers and developers who require automated research, citation management, and AI-powered assistance in a command-line environment.
- VibeKitvibekit.botVibeKit is a platform for developers to deploy persistent AI agents for their apps, offering integrated hosting, database, and DevOps automation. It supports web, iOS, API, and CLI interfaces, with features like GitHub integration, BYOK models, and auto-healing. Ideal for developers building and managing AI-powered applications.
- vibeissuepypi.orgvibeissue is a stdlib Python CLI tool that connects AI coding agents (such as those from Cursor or Claude) to GitHub Issues. It enables 'vibe coding' workflows by wiring autonomous agents directly to issue tracking. Distributed as a PyPI package under the MIT license with a GitHub repository, it is designed for developers integrating AI into their issue-driven development process.
- Void-Boxvoidplatform.aiVoid-Box provides a runtime environment for executing AI agents within disposable microVMs, focusing on strong isolation and explicit control over agent capabilities. It addresses the security challenges posed by autonomous agents that can execute code, access files, call APIs, and interact with external systems, emphasizing the need for containment and observability rather than relying on trust or prompt discipline. Each agent run is confined to its own microVM, with no shared kernel or ambient access to the host, ensuring that potential incidents are contained and do not compromise the broader environment. The platform allows users to wrap existing CLI agents or run remote agents within the same isolated runtime. Before execution, users declare the capabilities, resource limits, and allowed tools for each agent, such as network access, memory allocation, and specific skills. The runtime enforces these boundaries, and only explicitly permitted commands, files, and network access are available during execution. Output, logs, and artifacts from each run are preserved and remain inspectable after the VM exits, supporting traceability and auditability. Built-in observability features include structured logs, traces, metrics, and telemetry, with per-stage spans and event correlation for detailed inspection of agent behavior. Void-Box can be run locally or remotely, and supports Linux (KVM) and macOS (Apple Silicon) platforms. It is compatible with OCI images for reproducible root filesystems and supports interactive PTY shells for CLI agents. The platform is vendor-neutral by design and can restore microVMs in sub-second timeframes, enabling efficient snapshots and skipping full VM boots. The operator layer, Void-Control, coordinates multiple Void-Box runs and allows for inspection across stages. 0 license. It is intended for users who require secure, auditable execution of AI agents with explicit containment and runtime observability, such as developers and researchers working with autonomous systems.
- Vibe Modevibemode.aiVibe Mode is a native integrated development environment (IDE) designed specifically for AI-assisted coding on macOS. Built from the ground up without relying on Electron or existing code editors, it aims to deliver a fast and lightweight experience, with a download size of 100 MB and idle memory usage under 100 MB. The application is tailored for developers seeking an environment optimized for both performance and AI integration, emphasizing minimal resource consumption and rapid launch times of less than one second. The platform integrates a native inference engine that allows users to run AI models locally on Apple Silicon, achieving response times between 3 and 20 milliseconds to the first token. This local inference is handled by VibeMLX, a 17 MB Zig binary capable of loading HuggingFace models directly and running on Apple's Metal framework. The inference engine is compiled directly into the app's daemon, eliminating the need for Python runtimes or external dependencies. Users can also access AI models from a broad range of providers, including Anthropic, OpenAI, Google, xAI, Groq, DeepSeek, OpenRouter, Together, Cerebras, and others, as well as local models via VibeMLX, Ollama, or LM Studio, with no vendor lock-in. Vibe Mode's Samsara context engine indexes entire projects within milliseconds and maintains semantic continuity across sessions, allowing the AI assistant to understand and reference the full codebase and ongoing conversations. This enables features such as project-wide code understanding, advanced search, and persistent context for discussions and debugging. Security and privacy are prioritized, with API keys stored in the macOS Keychain, local inference that does not transmit code externally, and project-level isolation by design. The IDE also features a fully integrated, GPU-accelerated Ghostty terminal emulator, natively embedded within the application rather than as a subprocess. Vibe Mode includes a custom implementation of Jinja2 templating in Zig for native speed and compatibility. The tool is currently available as a private beta for macOS users, with access managed via a waitlist. Vibe Mode is developed by gremlinlabs.
- MimoBoxmimobox.ioMimoBox is an open-source CLI tool that provides a local sandbox runtime for AI agents. It allows developers to safely execute and test agents in an isolated environment, supporting secure experimentation and debugging.
- Open Vibeopenvibe.shOpen Vibe is an open-source platform designed to assist users in building and understanding production-ready SaaS applications with the help of AI agents such as Claude Code. The tool aims to bridge the gap between learning web development through static tutorials and building applications with AI, enabling users to gain a deeper understanding of the systems behind their SaaS projects while actively developing their own apps. The platform operates by running locally on the user's machine, ensuring that all code and processes remain under the user's control without platform lock-in or token-limit issues. Open Vibe provides a guided, interactive learning experience where an AI agent tutors the user step-by-step through the process of app development. Users interact with live diagrams overlaid on their running applications, receive explanations of backend processes, and can ask questions to clarify concepts as they build. The tool supports both beginners and experienced developers, requiring no prior coding knowledge, and allows users to progress at their own pace through modular phases. These phases include learning the basics of web apps, deploying full-stack applications, and building SaaS products using the included Open SaaS starter template, which features components like authentication, payments, admin, email, and AI integration. Open Vibe is delivered as a set of instructions and interactive components that integrate with AI agents capable of reading files and executing tools. Users can get started by pasting a provided prompt into their agent, which then guides them through the setup and development process. The course pack is provided in plain Markdown, making it compatible with various agents beyond Claude Code. The tool is MIT licensed, 100% free, requires no signup, and encourages forking, remixing, and even running classes with its materials. By combining hands-on building with real-time system explanations, Open Vibe targets aspiring founders and developers who want to both launch SaaS products and understand the underlying architecture, promoting learning and practical application in tandem.
- Vibe-Logvibe-log.devVibe-Log is a tool for understanding AI-assisted coding patterns in Claude Code and Codex. It focuses on where time goes during sessions, what actually gets shipped, and how prompting habits and work patterns can be improved. The site describes it as helping builders move from vibe-coding to more productive AI-driven coding. Its features center on session analysis and guidance. Vibe-Log offers real-time analytics for AI-assisted coding sessions, shows when a user is in flow or stuck, and provides strategic feedback inside the coding flow. It tracks features completed, bugs fixed, problems solved, and time breakdowns, and it surfaces peak performance windows, flow states, and triggers. The product also includes AI coaching with selectable coach personalities, personalized recommendations, AI efficiency scoring, and pattern recognition. Other named features include local HTML reports, detailed charts and metrics, weekly activity heatmaps, goal completion tracking, streaks, achievements, and auto-generated summaries that can be shared on X. The tool is delivered through a CLI command, with the shown command npx vibe-log-cli@latest and a note that it starts with a single command and no installation. It works with Claude Code and Codex sessions, and the page says it can run completely offline with data staying local. The page also describes local and cloud versions, both completely free. The local version runs on the user’s machine, is private and offline, and provides local HTML reports. The cloud version is presented as free forever and includes advanced AI insights, auto sync, daily summaries, reminders, and a gamified experience. The page also states that Vibe-Log is fully open source and emphasizes transparency and data ownership. It is described as useful for solo developers and for people building in public.
- vibepodvibepod.devvibepod is a command-line interface designed to simplify the use of containerized AI coding agents. It provides a single CLI for launching, managing, and interacting with AI agents that assist with software development tasks inside isolated containers.
- VibeFramevibeframe.aiVibeFrame is a storyboard-first video CLI for AI agents. It turns a written brief into project files, generated assets, scene composition, build reports, render inspection, and a final rendered MP4, with the CLI acting as the agent workflow layer while the coding agent remains the outer loop. The documented workflow starts from a rough brief in brief.md and optional files in a media/ folder. VibeFrame creates STORYBOARD.md and DESIGN.md, then provides commands for storyboard validation, planning, building, inspecting the project or render, repairing scenes, and rendering output. The page also describes lower-level media, scene, timeline, and YAML commands for debugging specific stages, plus escape-hatch paths for generating a standalone image, video, narration, music, or motion asset, and for editing or remixing existing media with captions, reframe, highlights, overlays, background music, and cleanup. It is intended for bash-capable coding agents and is described as CLI-first rather than terminal-only. The page names use with agents and AI apps such as OpenAI Codex, Claude, Cursor, Aider, Gemini CLI, OpenCode, and Claude Desktop. It also says the CLI can be used from a terminal, and that a desktop extension is available through a downloaded .mcpb file for opening a workspace folder without a terminal. Installation is shown with a shell command that fetches install.sh, and project setup writes ./.vibeframe/config.yaml for the workspace. The tool includes commands for setup, init, storyboard validation, planning, build dry runs with a max-cost gate, status, inspect, render, and scene repair. Its outputs include STORYBOARD.md, DESIGN.md, build-report.json, review-report.json, and JSON output from commands. The page also mentions host setup and doctor commands for integration configuration, and it presents an install script and GitHub documentation links.
Ranked by how close each one sits to VibeBox in the index, not by popularity. Back to VibeBox →