Tools Directory

Best AI coding tools list

A curated directory of AI IDEs, terminal agents, PR agents, open-source coding assistants, and MCP-ready developer workflows.

★ 4.8AI IDE · Codebase Chat · Agent Mode

Cursor

AI-first code editor for fast feature work

Cursor is a VS Code-based AI editor with native chat, tab completion, codebase context, and agent mode for multi-file changes.

IDE: Cursor editor, based on VS Code

Agent mode: Built-in agent mode for planning and editing across files

MCP fit: Supports MCP-style context workflows and pairs naturally with filesystem, GitHub, docs, browser, and database servers

Feature implementationRefactoringRepo Q&A

Read review →

★ 4.7Terminal Agent · Repo Automation · Test Runner

Claude Code

Terminal coding agent for codebase exploration and verified changes

Claude Code is an agentic terminal workflow for exploring repositories, editing files, running commands, and producing evidence-backed implementation reports.

IDE: Terminal-first; works alongside any editor

Agent mode: Strong multi-step terminal agent for inspect-edit-run-report loops

MCP fit: Can benefit from MCP servers for GitHub, browser QA, docs, databases, and deployment evidence

Repo onboardingBug fixingTest-backed implementation

Read review →

★ 4.6IDE Extension · GitHub Native · Pair Programmer

GitHub Copilot

Mainstream AI pair programmer across IDEs

GitHub Copilot provides inline completions, Copilot Chat, pull request assistance, and GitHub-native context across popular editors.

IDE: VS Code, Visual Studio, JetBrains, Neovim, and GitHub surfaces

Agent mode: Agentic assistance is improving across chat, workspace, and PR workflows

MCP fit: Best paired with GitHub, docs, and workflow MCP servers around team processes

AutocompletePR assistanceEveryday coding

Read review →

★ 4.7AI IDE · Autonomous Agent · Cascade

Windsurf

Agentic IDE with Cascade for autonomous edits

Windsurf is an agentic IDE from Codeium with Cascade, code completion, web-aware assistance, and multi-step coding task support.

IDE: Windsurf IDE

Agent mode: Cascade supports multi-step autonomous coding inside the editor

MCP fit: Useful with MCP-backed browser, GitHub, filesystem, and deployment checks

Autonomous implementationRefactoringRepo exploration

Read review →

★ 4.5Coding Agent · PR Workflow · Automation

OpenAI Codex

Cloud and CLI coding agent for PR-oriented work

OpenAI Codex workflows focus on delegating implementation, code review, tests, and pull-request style development to an agentic coding system.

IDE: CLI and cloud/PR-oriented workflows rather than one fixed IDE

Agent mode: Agentic delegation for implementation, tests, and review loops

MCP fit: Pairs with MCP context for repo, issue, docs, browser QA, and release workflows

PR reviewScoped feature workTest fixing

Read review →

★ 4.4Autonomous Agent · Cloud Workspace · Engineering Tasks

Devin

Autonomous software engineering agent for delegated tasks

Devin is an autonomous AI software engineer designed to take larger engineering tasks, work through repos, use tools, and return completed implementation attempts.

IDE: Cloud agent workspace rather than a daily IDE replacement

Agent mode: High-autonomy software engineering agent with planning and execution loops

MCP fit: Use MCP thinking around least-privilege repo, issue, browser, and deployment access when designing workflows

Delegated ticketsInternal toolsRepo tasks

Read review →

★ 4.3Hosted IDE · App Builder · Deployment

Replit Agent

Prompt-to-app coding agent inside Replit

Replit Agent helps builders create, modify, and deploy applications inside Replit from natural-language prompts and iterative app-building flows.

IDE: Replit browser IDE and hosted workspace

Agent mode: Prompt-to-app agent that can create files, modify code, and help deploy

MCP fit: Most useful when connected workflows define external APIs, docs, GitHub, and deployment boundaries

Prototype shippingInternal demosLearning projects

Read review →

★ 4.2Agent Platform · Engineering Automation · Team Workflow

Factory

Agentic development platform for engineering workflows

Factory focuses on AI agents for software development workflows, helping engineering teams automate coding tasks, reviews, and delivery operations.

IDE: Platform-oriented workflow; integration surface depends on setup

Agent mode: Agentic workflows for coding, review, and delivery operations

MCP fit: Maps well to MCP governance for internal tools, repo systems, and delivery pipelines

Engineering automationReview workflowsTeam agents

Read review →

★ 4.4Open Source · VS Code · Tool Use

Cline

Open-source VS Code coding agent with tool use

Cline is an open-source coding agent for VS Code that can inspect projects, edit files, run terminal commands, use browser automation, and work with multiple model providers.

IDE: VS Code extension

Agent mode: Agentic tool use with file edits, terminal commands, and browser actions

MCP fit: Strong fit for MCP because tool use, permissions, and external context are central to the workflow

VS Code agent workTool-assisted debuggingLocal repo edits

Read review →

★ 4.4Open Source · Terminal · Git Workflow

Aider

Open-source AI pair programming in the terminal

Aider is an open-source terminal pair programmer that edits files in a git repo and works with multiple LLM providers.

IDE: Terminal-first; works with any editor

Agent mode: Pair-programming loop with file edits rather than full autonomous workspace by default

MCP fit: Can be combined with MCP-style scripts and tools around docs, GitHub, tests, and release steps

Small repo editsScript changesCLI-first development

Read review →

★ 4.3Open Source · IDE Extension · Customizable

Continue

Open-source AI coding assistant for VS Code and JetBrains

Continue is an open-source AI coding assistant that lets teams customize models, context providers, prompts, and IDE workflows.

IDE: VS Code and JetBrains extensions

Agent mode: Assistant workflow is highly configurable; deeper agentic behavior depends on setup

MCP fit: Natural fit for MCP because configurable context and tool access are core design concerns

Private code assistanceInternal AI platformCustom model rollout

Read review →

★ 4.2Open Source · Autonomous Agent · Workspace Automation

OpenHands

Open-source software engineering agent for autonomous repo work

OpenHands is an open-source agentic coding platform for delegating software engineering tasks, inspecting repositories, editing files, running commands, and iterating inside controlled workspaces.

IDE: Workspace and platform workflow rather than a single IDE extension

Agent mode: Autonomous software engineering agent with planning, file edits, commands, and iteration loops

MCP fit: Strong fit for MCP-style governance because external tools, repo access, browser checks, and deployment actions need clear boundaries

Autonomous repo tasksAgent platform experimentsInternal tooling

Read review →

★ 4.5Code Security · Static Analysis · CI Review

Semgrep

Code security and review automation for AI-assisted teams

Semgrep scans source code for security issues, policy violations, and risky patterns, making it useful as a review gate around AI-generated code.

IDE: CLI, CI, GitHub workflows, and editor integrations depending on setup

Agent mode: Not an autonomous coding agent; best used as a guardrail that reviews agent output

MCP fit: Useful alongside GitHub, filesystem, CI, and security evidence MCP workflows

AI code securityPull request reviewCI gates

Read review →

★ 4.4Code Security · SAST · Dependency Security

Snyk Code

Developer-first security scanning for code and dependencies

Snyk Code and the broader Snyk platform help teams find vulnerabilities in source code, dependencies, containers, and infrastructure before AI-assisted changes ship.

IDE: IDE plugins, CLI, SCM, and CI integrations depending on setup

Agent mode: Not a coding agent; use as a security review system around agent-generated pull requests

MCP fit: Pairs with GitHub, dependency, CI, and security-report workflows where agents need evidence rather than broad secrets

Secure code reviewDependency checksPR scanning

Read review →

★ 4.5Code Security · Semantic Analysis · GitHub Native

GitHub CodeQL

Semantic code analysis for GitHub security review

GitHub CodeQL analyzes code with semantic queries to find vulnerabilities and security patterns, especially inside GitHub-native pull request and code scanning workflows.

IDE: GitHub code scanning, Actions, CLI, and extension workflows depending on setup

Agent mode: Not a coding agent; use as an automated reviewer and security gate for agent-created changes

MCP fit: Natural fit with GitHub MCP workflows for pull requests, code scanning alerts, and evidence collection

GitHub security reviewPR gatesCritical code analysis

Read review →

★ 4.3Code Review · GitHub Native · PR Workflow

GitHub Copilot Code Review

AI-assisted pull request review inside GitHub

GitHub Copilot code review features help identify issues, suggest improvements, and accelerate pull request feedback inside GitHub workflows.

IDE: GitHub pull requests and related Copilot surfaces

Agent mode: Review assistant rather than a full autonomous implementation agent

MCP fit: Pairs with GitHub, CI, docs, and security MCP workflows for evidence-backed review loops

Pull request reviewTeam feedbackAI-generated code checks

Read review →

★ 4.2Terminal Agent · Google AI · Large Context

Gemini CLI

Terminal AI agent for Google model workflows and large-context coding

Gemini CLI brings Google Gemini models into command-line development workflows for repo exploration, implementation support, debugging, and long-context research around codebases.

IDE: Terminal-first; works alongside any editor

Agent mode: CLI assistant and agent-style loop depending on configuration and permissions

MCP fit: Useful with MCP-backed docs, GitHub, filesystem, browser, and deployment evidence workflows

Large-context repo researchCLI-assisted implementationDebugging

Read review →

★ 4.1Coding Agent · Open Model · Terminal Workflow

Qwen Code

Open coding-agent workflow built around Qwen model access

Qwen Code is a coding-agent workflow for developers who want to use Qwen models for repository understanding, file edits, command execution, and coding task assistance.

IDE: Terminal and integration-dependent workflows rather than one fixed IDE

Agent mode: Agentic behavior depends on selected client, tools, and execution sandbox

MCP fit: Strong fit for MCP-style tool boundaries when using open models with filesystem, GitHub, and docs context

Model experimentationCLI codingMultilingual development

Read review →

★ 4.3Open Source · VS Code · Tool Use

Roo Code

Open-source VS Code agent for structured coding modes

Roo Code is an open-source VS Code coding agent that supports multiple task modes, provider choices, file edits, terminal actions, and controlled tool use inside developer workspaces.

IDE: VS Code extension

Agent mode: Agentic tool use with configurable modes for code, architecting, debugging, and review

MCP fit: Good fit for MCP because permissions, tools, and provider routing are central to the workflow

VS Code agent workMode-based implementationLocal debugging

Read review →

★ 4.2Codebase Chat · Enterprise Search · IDE Extension

Sourcegraph Cody

Codebase-aware AI assistant for enterprise repositories

Sourcegraph Cody helps developers ask questions about large codebases, generate edits, and use Sourcegraph code intelligence across repositories and enterprise development environments.

IDE: VS Code, JetBrains, and Sourcegraph surfaces depending on setup

Agent mode: Assistant and edit workflow rather than high-autonomy task delegation

MCP fit: Pairs with docs, GitHub, and internal knowledge MCP workflows for large-codebase onboarding

Large codebase Q&ARepo onboardingCross-repo search

Read review →

★ 4.1Code Completion · Enterprise · IDE Extension

Tabnine

Enterprise AI code completion with privacy controls

Tabnine provides AI code completion and chat across developer IDEs with a focus on enterprise privacy, deployment control, and team-wide coding assistance.

IDE: VS Code, JetBrains, Visual Studio, Eclipse, and other IDE integrations depending on current support

Agent mode: Primarily assistant and completion workflow rather than autonomous coding agent

MCP fit: Best paired with MCP-governed review, docs, and security workflows when teams need more than autocomplete

AutocompleteEnterprise rolloutEveryday coding

Read review →