What is Crush?

Crush is an advanced AI coding assistant that operates directly within your terminal, seamlessly connecting your tools, code, and workflows with the large language model (LLM) of your choice. It offers a versatile model selection, enabling users to choose from an array of LLMs or to implement their own through APIs compatible with OpenAI or Anthropic, while also allowing for mid-session changes between models without losing context. Built with session-based functionality in mind, Crush supports multiple project-specific contexts running concurrently. With enhancements from Language Server Protocol (LSP), it delivers coding-aware context akin to that found in popular developer editors, elevating the coding experience. The tool boasts high customizability through Model Context Protocol (MCP) plugins, which can be utilized via HTTP, stdio, or SSE to broaden its functionalities. Crush can run on any operating system, utilizing Charm’s refined Bubble Tea-based terminal user interface for an elegant experience. Developed in Go and available under the MIT license (with FSL-1.1 for trademark considerations), Crush allows developers to work within their terminal while enjoying sophisticated AI coding assistance, significantly optimizing their workflows. Its groundbreaking design not only boosts productivity but also fosters a smooth integration of AI into the daily routines of programmers, making coding more efficient and enjoyable than ever before. Moreover, the continuous evolution of its features ensures that users will always have access to the latest advancements in AI-assisted coding.

Pricing

Price Starts At:
Free
Free Version:
Free Version available.

Integrations

Offers API?:
Yes, Crush provides an API
All Crush Integrations
Similar Software to Crush
Google AI Studio Reviews & Ratings
Google AI Studio
(30 Ratings)
Google AI Studio is a comprehensive platform for discovering, building, and operating AI-powered applications at scale. It unifies Google’s leading AI models, including Gemini 3.5, Imagen, Veo, and Gemma, in a single workspace. Developers can test and refine prompts across text, image, audio, and video without switching tools. The platform is built around vibe coding, allowing users to create applications by simply describing their intent. Natural language inputs are transformed into functional AI apps with built-in features. Integrated deployment tools enable fast publishing with minimal configuration. Google AI Studio also provides centralized management for API keys, usage, and billing. Detailed analytics and logs offer visibility into performance and resource consumption. SDKs and APIs support seamless integration into existing systems. Extensive documentation accelerates learning and adoption. The platform is optimized for speed, scalability, and experimentation. Google AI Studio serves as a complete hub for vibe coding–driven AI development.
Learn more
JetBrains Junie Reviews & Ratings
JetBrains Junie
(12 Ratings)
Junie, the AI coding agent by JetBrains, revolutionizes the way developers interact with their code by embedding intelligent assistance directly into JetBrains IDEs like WebStorm, RubyMine, and GoLand. Designed to fit naturally into developers’ existing workflows, Junie helps tackle both small and ambitious coding tasks by providing tailored execution plans and automated code generation. It combines the power of AI with IDE capabilities to perform code inspections, syntax checks, and run tests automatically, maintaining code quality without manual intervention. Junie offers two distinct modes: one for executing code tasks and another for interactive querying and planning, allowing developers to seamlessly collaborate with the agent. Its ability to comprehend code relationships and project logic enables it to propose efficient solutions and reduce time spent on debugging. Developers from various fields, including game development and web design, have showcased impressive projects built entirely or partly with Junie’s assistance. The tool supports multi-file edits and integrates version control system (VCS) assistance, making complex refactoring easier and safer. JetBrains offers multiple pricing plans tailored to individuals and organizations, ranging from free tiers to premium AI Ultimate for intensive daily use. By handling repetitive coding chores, Junie frees developers to focus on the creative and strategic aspects of software development. Overall, Junie stands as a powerful AI companion transforming traditional coding into a smarter, more collaborative experience.
Learn more
Claude Code Reviews & Ratings
Claude Code
(1 Rating)
Claude Code is an advanced AI coding assistant created to deeply understand and work within real software projects. Unlike traditional coding tools that focus on syntax or snippets, it comprehends entire repositories, dependencies, and architecture. Developers can interact with Claude Code directly from their terminal, IDE, Slack workspace, or the web interface. By using natural language prompts, users can ask Claude to explain unfamiliar code, refactor components, or implement new features. The tool performs agentic searches across the codebase to gather context automatically, removing the need to manually select files. This makes it especially valuable when joining new projects or working in large, complex repositories. Claude Code can also run CLI commands, tests, and scripts as part of its workflow. It integrates with version control platforms to help manage issues, commits, and pull requests. Teams benefit from faster iteration cycles and reduced context switching. Claude Code supports multiple powerful Claude models depending on the plan selected. Usage scales from short sprints to large, ongoing development efforts. Overall, it acts as a collaborative coding partner that enhances productivity without disrupting established workflows.
Learn more
Muse Code Reviews & Ratings
Muse Code
(1 Rating)
Muse Code is Meta’s terminal-based AI coding agent built to take on complex software engineering tasks across large repositories. The agent is powered by Muse Spark 1.2 and is designed to plan code changes, write implementation code, validate results, and support end-to-end developer workflows. Muse Code can coordinate multiple persistent subagents for each task, helping solve difficult problems faster and with less manual intervention. Its architecture uses a simple agent loop enhanced by async background agents that remain active for the full session instead of being spawned only for individual steps. These background agents reduce repeated information gathering, carry out next actions, and communicate back to the main agent when useful. Muse Code’s runtime uses a local event log where model calls, tool runs, approvals, and edits are continuously appended. This event log serves as a single source of truth, making the runtime replay-exact and restart-safe if a crash or interruption occurs. The design allows Muse Code to handle long-running development work without losing progress or context. Muse Code includes bundled skills such as /plan for approval-gated task planning, /grill for stress-testing plans, and /goal for working toward successful completion of a defined objective. Example workflows include interpreting a video input, understanding the requested output, and producing a rich software experience such as a vacation home marketing and booking page. By combining terminal execution, autonomous planning, persistent background agents, replay-safe runtime design, bundled skills, and Muse Spark 1.2 model support, Muse Code helps developers complete ambitious coding tasks with greater reliability.
Learn more

Screenshots and Video

Play Video Crush Screenshot 1 Crush Screenshot 2

Company Facts

Company Name:
Charm
Date Founded:
2019
Company Location:
United States
Company Website:
github.com/charmbracelet/crush

Product Details

Deployment
Windows
Mac
Linux
Training Options
Documentation Hub
On-Site Training
Video Library
Support
Standard Support
Web-Based Support

Product Details

Target Company Sizes
Individual
1-10
11-50
51-200
201-500
501-1000
1001-5000
5001-10000
10001+
Target Organization Types
Mid Size Business
Small Business
Enterprise
Freelance
Nonprofit
Government
Startup
Supported Languages
English

Crush Categories and Features

Agentic CLI Coding Tools

Agentic AI Platform

AI Coding Agents

Crush Customer Reviews

Write a Review
  • Reviewer Name: A Verified Reviewer
    Position: Developer
    Has used product for: Less than 6 months
    Uses the product: Daily
    Org Size (# of Employees): 5,000 - 9,999
    Feature Set
    Layout
    Ease Of Use
    Would you Recommend to Others?
    1 2 3 4 5 6 7 8 9 10

    Crush AI coding agent review

    Date: Jul 31 2026
    Summary

    Five stars from me. Crush feels like a strong AI coding agent for developers who want something terminal-native, open source, customizable, and close to their real workflow.

    It is not a magic autopilot, and I would not run it recklessly. But as a supervised coding partner for debugging, refactoring, repo exploration, command-line workflows, and agentic development, Crush looks like one of the more interesting tools in the AI coding agent space.

    Positive

    Crush is really appealing from a developer’s point of view because it brings agentic coding into the terminal without making the workflow feel bloated. It feels like a tool for people who already live in the command line and want an AI agent that can work with their existing code, tools, and habits.

    I like that it is open source and model-flexible. Being able to wire your own LLM into the workflow is a big deal because different coding tasks need different tradeoffs around cost, speed, context, and reasoning quality.

    The terminal UI is also a nice fit for real development work. Crush is not trying to replace the whole IDE; it gives you an agentic layer where you can ask for help, work through code, run commands, and stay close to the environment you already use.

    The active GitHub repo is a good sign too. Frequent commits, lots of stars, and ongoing work around providers, MCP, hooks, and shell features make it feel like a project with real momentum instead of a one-off AI demo.

    Negative

    Crush is probably not the best fit for someone who wants a polished, hand-holdy coding assistant with everything abstracted away. It feels more geared toward developers who are comfortable in the terminal and want control over models, config, permissions, and workflows.

    I would also be careful with any AI coding agent that can interact with your shell and project files. Recent research on coding-agent setup workflows shows that agents can be vulnerable to malicious or misleading project instructions, dependency names, registry redirects, and unsafe install steps, so developers still need review habits and guardrails.

    For production repos, I would keep permissions tight, review commands, watch file diffs, and avoid letting any agent run destructive operations without explicit approval.

    Read More...
  • Previous
  • You're on page 1
  • Next