coding-agents-matrix

KO | EN

gitlite — search

#python #java #python3 #arduino #golang #machine-learning #rust #html #flask #javascript #seismology #nodejs

coding-agents-matrix

★ 19 Open GitHub ↗

A curated comparison matrix of ai coding agents

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# AI Coding Agents Matrix - Best AI Coding Agents List

The most comprehensive **AI coding agents list** for developers. Compare **24+ best AI coding agents** side-by-side including Claude Code, Cursor, GitHub Copilot, Aider, Windsurf, Cline, Gemini CLI, and more.

Whether you're looking for the best AI code assistant for your workflow, comparing open source vs proprietary options, or evaluating features like MCP support, custom rules, and plan mode - this matrix helps you make an informed decision.

**Live Demo:** [https://coding-agents-matrix.dev](https://coding-agents-matrix.dev)

Powered by [Packmind](https://packmind.com).

## Table of Contents

- [Features](#features)
- [Getting Started](#getting-started)
- [Adding a New Agent](#adding-a-new-agent)
- [Agent Properties Reference](#agent-properties-reference)
- [Property Values](#property-values)
- [Examples](#examples)
- [Contribute](#contribute)

## What is an AI Coding Agent?

An **AI coding agent** is an autonomous software tool that uses large language models (LLMs) to help developers write, edit, debug, and refactor code. Unlike simple code completion tools, AI coding agents can:

- Understand your entire codebase context
- Execute multi-step tasks autonomously
- Run commands, tests, and builds
- Create and modify multiple files
- Integrate with external tools via protocols like MCP

The best AI coding agents in 2026 offer features like **plan mode** (planning before execution), **sub-agents** (delegating subtasks), **custom rules** (project-specific instructions), and **BYO LLM** (bring your own language model).

## AI Coding Agents Included

This list tracks the best AI coding agents available in 2026:

| Agent | Type | Notable Features |
|-------|------|------------------|
| **Aider** | Open Source | Git integration, voice input, BYO LLM |
| **Amp** | Proprietary | Subagents, code review, issue integration |
| **Antigravity** | Proprietary | Google's AI IDE, Gemini-powered |
| **AugmentCode** | Proprietary | Slack/Teams integration, smart hooks |
| **Claude Code** | Proprietary | MCP support, hooks, sub-agents, plan mode |
| **Cline** | Open Source | VS Code extension, transparent consumption |
| **Continue** | Open Source | Mission Control UI, cloud agent support |
| **Crush** | Open Source | TUI interface, multi-model configs |
| **Cursor** | Proprietary | VS Code fork IDE, background agents |
| **Factory** | Proprietary | Droids, memory, Jira/Notion integration |
| **Gemini CLI** | Open Source | 1000 free requests/day, LSP integration |
| **GitHub Copilot** | Proprietary | BYOK for enterprise, memory features |
| **Goose** | Open Source | Recipes, Agent Client Protocol |
| **Jules** | Proprietary | GitHub integration, async cloud sandbox |
| **KiloCode** | Open Source | Multi-mode context, Memory Bank |
| **Kiro** | Proprietary | AWS integration, spec-driven development |
| **Mistral Vibe CLI** | Open Source | Devstral models, ACP support |
| **OpenAI Codex** | Proprietary | Custom prompts, skills |
| **OpenCode** | Open Source | TUI, ACP support, plugins |
| **RooCode** | Open Source | VS Code extension, cloud agents |
| **Trae AI** | Open Source | Bytedance, multi-agent orchestration |
| **Warp** | Proprietary | GPU-accelerated terminal, BYOK |
| **Windsurf** | Proprietary | Cascade agent, worktrees |
| **Zed** | Open Source | Fast editor, ACP support |

## Features

- **Interactive Table**: Sort by any column to compare agents
- **Advanced Filtering**: Search and filter by type, features, and capabilities
- **Beautiful Design**: Modern UI with smooth animations and hover effects
- **Responsive**: Works perfectly on desktop, tablet, and mobile devices
- **Visual Badges**: Color-coded badges for easy capability identification

### Sorting

Click any column header to sort:
- First click: Ascending order
- Second click: Descending order
- Third click: Return to default order

### Filtering

- **Search**: Filter by agent name, type, or description
- **Type**: Filter by Open Source or Proprietary
- **Features**: Select from multiple filters including CLI, Dedicated IDE, IDE Extension, BYO LLM, MCP Support, Custom Rules, AGENTS.md, Skills, Commands, Subagents, and Plan Mode

All filters use AND logic for precise results.

## Getting Started

### Installation

```bash
npm install
```

### Development

```bash
npm run dev
```

Visit `http://localhost:5173` to see the application.

### Build

```bash
npm run build
```

### Preview Production Build

```bash
npm run preview
```

## Adding a New Agent

To add a new agent to the board, follow these steps:

1. **Open the agents data file**: `/src/data/agents-detailed.json`

2. **Add a new agent object** to the array with all required properties:

```json
{
  "name": "Your Agent Name",
  "type": "Open Source",
  "github": { "value": "https://github.com/owner/repo", "detail": null },
  "ghStars": { "value": null, "detail": null },
  "website": { "value": "https://youragent.com", "detail": null },
  "firstRelease": { "value": "2025-01", "detail": null },
  "cli": { "value": "yes", "detail": null },
  "dedicatedIde": { "value": "no", "detail": null },
  "ideExtension": { "value": "no", "detail": null },
  "byoLlm": { "value": "yes", "detail": "Supports OpenAI and Anthropic" },
  "mcpSupport": { "value": "yes", "detail": null },
  "customRules": { "value": "yes", "detail": null },
  "agentsMdSupport": { "value": "yes", "detail": null },
  "agentSkillsSupport": { "value": "yes", "detail": null },
  "commandsReusablePrompts": { "value": "yes", "detail": null },
  "subagentsSupport": { "value": "no", "detail": null },
  "planMode": { "value": "yes", "detail": null },
  "additionalInfo": "Brief description of unique features"
}
```

**Note about the `detail` field:**
- Use it to provide additional context that will appear as a tooltip in the UI
- Set to `null` when no clarification is needed
- Examples: links to documentation, clarifications about partial support, or caveats

3. **Save the file** - the UI will automatically pick up the new agent

## Agent Properties Reference

### Data Structure

All properties except `name`, `type`, and `additionalInfo` use this structure:
```json
{
  "propertyName": {
    "value": "yes" | "no" | "partial" | string | number | null,
    "detail": "Optional context displayed as tooltip in UI"
  }
}
```

### Property Reference

| Property | Type | Values | Description |
|----------|------|--------|-------------|
| **Identity** | | | |
| `name` | string | - | Official agent name |
| `type` | string | `"Open Source"` \| `"Proprietary"` | Licensing model |
| `github` | object | `{value: "url" \| null, detail: null}` | GitHub repository URL |
| `ghStars` | object | `{value: number \| null, detail: "Last updated: date"}` | Star count (auto-updated) |
| `website` | object | `{value: "url" \| null, detail: null}` | Official website URL |
| `firstRelease` | object | `{value: "YYYY-MM" \| null, detail: null}` | Initial release date |
| **Packaging** | | | |
| `cli` | object | `{value: "yes" \| "no" \| "partial", detail: string \| null}` | CLI support |
| `dedicatedIde` | object | `{value: "yes" \| "no", detail: string \| null}` | Standalone/specific IDE |
| `ideExtension` | object | `{value: "yes" \| "no", detail: string \| null}` | Available as extension |
| **Features** | | | |
| `byoLlm` | object | `{value: "yes" \| "no" \| "partial", detail: string \| null}` | Bring Your Own LLM |
| `mcpSupport` | object | `{value: "yes" \| "no", detail: string \| null}` | [Model Context Protocol](https://modelcontextprotocol.io/) |
| `customRules` | object | `{value: "yes" \| "no", detail: string \| null}` | Custom instructions |
| `agentsMdSupport` | object | `{value: "yes" \| "no" \| "partial", detail: string \| null}` | AGENTS.md file support |
| `agentSkillsSupport` | object | `{value: "yes" \| "no", detail: string \| null}` | Skills/plugins system |
| `commandsReusablePrompts` | object | `{value: "yes" \| "no", detail: string \| null}` | Slash commands |
| `subagentsSupport` | object | `{value: "yes" \| "no" \| "partial", detail: string \| null}` | Sub-agent spawning |
| `planMode` | object | `{value: "yes" \| "no", detail: string \| null}` | Planning before execution |
| `additionalInfo` | string | - | Brief unique features |

### Detail Field Usage

The `detail` field provides context shown as tooltips:
- **Clarification**: "Terminal agent and VS Code plugin available"
- **Links**: Documentation URLs for features
- **Caveats**: "Not native, but can be configured..."
- **Null**: When no additional context needed

---

## Property Values

The UI renders different visual indicators based on property values:

| Value | Badge Color | Icon | Meaning |
|-------|-------------|------|---------|
| `"yes"` | Green | ✓ | Feature is fully supported |
| `"no"` | Red | ✗ | Feature is not available |
| `"partial"` or `"limited"` | Amber | ◯ | Feature is partially supported or has limitations |
| `null` | Gray | — | Not applicable or not yet determined |

## Examples

### Example 1: Complete Entry (Open Source Agent)

This example shows a complete agent entry using real data from Aider, demonstrating all properties with the correct `{value, detail}` structure:

```json
{
  "name": "Aider",
  "type": "Open Source",
  "github": {
    "value": "https://github.com/Aider-AI/aider",
    "detail": null
  },
  "ghStars": {
    "value": 39874,
    "detail": "Last updated: 2026-01-19"
  },
  "website": {
    "value": "https://aider.chat/",
    "detail": null
  },
  "firstRelease": {
    "value": "2023-06",
    "detail": null
  },
  "cli": {
    "value": "yes",
    "detail": null
  },
  "dedicatedIde": {
    "value": "no",
    "detail": null
  },
  "ideExtension": {
    "value": "no",
    "detail": null
  },
  "byoLlm": {
    "value": "yes",
    "detail": "Supports connecting to OpenAI, Anthropic or local models via API keys"
  },
  "mcpSupport": {
    "value": "no",
    "detail": "No built-in MCP client; can be extended via external MCP server integrations"
  },
  "customRules": {
    "value": "yes",
    "detail": "Through the usage of the file `CONVENTIONS.md`"
  },
  "agentsMdSupport": {
    "value": "partial",
    "detail": "Not native, but can be configured to read these files. Read more: https://aider.chat/docs/usage/conventions.html#always-load-conventions"
  },
  "agentSkillsSupport": {
    "value": "no",
    "detail": null
  },
  "commandsReusablePrompts": {
    "value": "no",
    "detail": "All instructions given via natural language chat"
  },
  "subagentsSupport": {
    "value": "no",
    "detail": null
  },
  "planMode": {
    "value": "no",
    "detail": "Has a `/architect` mode but not considered as a `plan` mode."
  },
  "additionalInfo": "Integrates with Git (auto-commits changes with messages); voice input and image context support"
}
```

### Example 2: Minimal Entry (Proprietary Agent)

This example shows a minimal entry for a proprietary agent without a GitHub repository:

```json
{
  "name": "Jules",
  "type": "Proprietary",
  "github": {
    "value": null,
    "detail": null
  },
  "ghStars": {
    "value": null,
    "detail": null
  },
  "website": {
    "value": "https://jules.google",
    "detail": null
  },
  "firstRelease": {
    "value": "2024-12",
    "detail": null
  },
  "cli": {
    "value": "yes",
    "detail": "Jules Tools CLI to interface with the cloud agent"
  },
  "dedicatedIde": {
    "value": "no",
    "detail": "Runs asynchronously in cloud; not inside your editor"
  },
  "ideExtension": {
    "value": "no",
    "detail": null
  },
  "byoLlm": {
    "value": "no",
    "detail": "Powered by Google's models on their backend"
  },
  "mcpSupport": {
    "value": "yes",
    "detail": "Jules uses tools in a sandboxed cloud environment to compile, test, etc., though not user-configurable MCP"
  },
  "customRules": {
    "value": "no",
    "detail": "No support for AGENTS.md or custom instructions files; agent infers context from repository code and standard config"
  },
  "agentsMdSupport": {
    "value": "no",
    "detail": null
  },
  "agentSkillsSupport": {
    "value": "no",
    "detail": null
  },
  "commandsReusablePrompts": {
    "value": "yes",
    "detail": "Tasks are queued via CLI commands or GitHub comments – e.g. you instruct Jules to fix tests and it works asynchronously"
  },
  "subagentsSupport": {
    "value": "no",
    "detail": "Jules itself is a single agent per task, operating asynchronously"
  },
  "planMode": {
    "value": "yes",
    "detail": "Creates execution plans before implementation"
  },
  "additionalInfo": "Works via GitHub integration; operates asynchronously in secure cloud sandbox"
}
```

## Contribute

We welcome contributions! Here are three ways you can help improve the coding-agents-matrix:

### 1. Add a New Agent

Use the **[Add New Agent](https://github.com/PackmindHub/coding-agents-matrix/issues/new?template=02-add-new-agent.md)** issue template to submit a new AI coding agent for inclusion in the comparison matrix.

**When to use:** You've discovered a new coding agent that should be tracked and compared with others.

**What you'll provide:**
- Agent name, type (Open Source/Proprietary), and website
- GitHub URL and first release date
- Support status for 11+ features (yes/no/partial/null)
- Brief description highlighting unique capabilities

### 2. Update an Existing Agent

Use the **[Update Existing Agent](https://github.com/PackmindHub/coding-agents-matrix/issues/new?template=01-update-agent.md)** issue template to propose updates to any agent's information.

**When to use:** An agent's properties have changed, new features are available, or information is outdated or incorrect.

**What you'll provide:**
- Agent name and brief description of changes
- Updated property values with supporting evidence
- Links to documentation or references

### 3. Propose a New Property

Use the **[Add New Property](https://github.com/PackmindHub/coding-agents-matrix/issues/new?template=03-add-new-property.md)** issue template to propose a new property or feature to track across all agents.

**When to use:** You think an important capability or characteristic should be added to the comparison matrix for all agents.

**What you'll provide:**
- Property name (technical name in camelCase)
- Property group (Identity/Packaging/Features)
- Clear description and motivation
- Examples of agents that support this property
- Value format (yes/no/partial/null or custom values)

### Tips for Contributing

- Review the [Agent Properties Terminology](#agent-properties-reference) section to understand the existing properties
- Check existing properties to avoid duplication
- Provide sources or references to support your changes
- Use clear, concise language in descriptions
- Verify your proposed values against official agent documentation

---

## Updating GitHub Stars

The following script:

```bash
npm run update-stars
```

is run weekly and updates automatically the values in the database file.

## Verification Checklist

After your contribution is accepted and merged:
1. Verify all required properties are present and correct
2. Use consistent terminology and values with existing agents
3. Keep `additionalInfo` concise and informative
4. Confirm the agent/property renders correctly in both table and card views
5. Test filtering and sorting with the new agent/property included

coding-agents-matrix

Explore Similar Repositories

Was this content helpful?

Recent Feedback

Download README