Best AI tools for code

Open-source AI pair programming in the terminal

AWS coding assistant and transformation agent

Agentic coding tool succeeding Cody

Google's agent-first IDE built on Gemini 3

Context-aware coding agent for large codebases

No-code AI app builder

AI code review agent and dev assistant

Code chat, completion and search assistant

AI code search; now legacy-code modernization

Prompt-to-full-stack web apps in the browser

Terminal-based agentic coding assistant

Open-source autonomous coding agent for VS Code

Free multilingual code assistant from Zhipu

AI coding agents platform and IDE extensions

AI code review on every pull request

Open-source, customizable AI code assistant

Prompt-to-app and site builder

AI-first code editor with agentic multi-file editing

AI developer that builds full-stack apps with you

Autonomous AI software engineer for delegated tasks

AI code reviews and automated fixes

Enterprise 'Droids' software engineering agents

Google's cloud AI workspace for full-stack apps

Open-source terminal AI agent

AI code completion and chat for IDEs and Google Cloud

AI pair programmer: completions, chat and coding agent in IDEs

AI features across the GitLab DevSecOps platform

Stacked PRs with Diamond AI review

AI reviewer with full-codebase context

AI chat and completion across JetBrains IDEs

Asynchronous coding agent that works on your repos

JetBrains' autonomous coding agent

Spec-driven agentic IDE from AWS

Text-to-app builder for non-coders

Frontier code models with 100M-token context (in development)

AI-maintained developer documentation

Mass-scale automated code refactoring

AI-driven end-to-end web testing

Cloud + CLI software engineering agent

Open-source platform for autonomous dev agents

AI answer engine for programming questions

On-device developer memory and snippet copilot

Foundation models for software engineering

All-in-one AI app builder (GPT Pilot)

AI code integrity: tests, reviews and PR agents

Open-source customizable coding agent

Open frontier models aimed at autonomous coding

Browser IDE with Agent that builds full apps from prompts

Open-source agentic coding extension (Cline fork)

Prompt-to-native-mobile-app builder

Codebase-aware AI assistant

Automated code review and refactoring for Python

Research agent that fixes GitHub issues autonomously

AI junior dev turning issues into PRs

AI documentation that stays in sync with code

Self-hosted open-source coding assistant

Privacy-focused AI code completion for enterprises

ByteDance's free AI-native IDE

Generative UI and app builder

AI-powered terminal / agentic dev environment

Agentic AI IDE (Cascade) from the former Codeium

High-performance editor with built-in AI agents

Managed access to foundation models on AWS

Scalable AI compute built on Ray

ML/LLM observability and evaluation

Microsoft's platform to build and run AI apps

Deploy and serve ML models in production

Open-source model serving and deployment

Evals and observability for AI products

Record-speed inference on wafer-scale chips

Open-source embedding database for AI apps

Specialized GPU cloud for AI at scale

Fast inference for generative media models

Fast, production-grade inference platform

Evaluation and observability for AI agents

Privacy-first local LLM desktop app

Ultra-fast LLM inference on custom LPUs

Open-source LLM observability and gateway

The hub for open models, datasets and ML tools

Open-source offline ChatGPT alternative

Open embeddings, rerankers and search infra

Data labeling and evaluation platform

GPU cloud and workstations for AI

Open-source multimodal/vector database

Open-source LLM engineering platform

Observability and evals for LLM apps

Call 100+ LLM APIs with one interface

Efficient LLM inference in C/C++

Ship an LLM as a single executable file

Desktop app to run local LLMs

Open-source vector database at scale (Zilliz)

Serverless cloud for AI/ML workloads

Full-stack AI cloud (European)

Run open LLMs locally with one command

Unified API and marketplace for 300+ models

Automated evaluation and guardrails for LLMs

Managed vector database for AI apps

AI gateway and observability for production

Open-source LLM testing and red-teaming

Prompt management and observability

High-performance open-source vector database

Accurate document parsing for AI pipelines

Run and deploy open models via API

Build and deploy computer-vision models

Affordable GPU cloud for AI

Enterprise AI chips and inference cloud

Data labeling and RLHF for AI models

Fast serving framework for LLMs/VLMs

Programmatic data labeling and curation

Premium RLHF data labeling

Fast inference and training for open models

Cost-efficient vector/search database

AGI training data and coding-expert services

Preprocess documents for LLMs/RAG

Platform to build and evaluate LLM apps

Google Cloud's end-to-end ML/GenAI platform

High-throughput LLM serving engine

State-of-the-art embedding and rerank models

Open-source AI-native vector database

Experiment tracking and LLMOps (Weave)

AI email security against advanced attacks

AI analyst for the Falcon platform

Self-learning AI for cyber defense

Autonomous AI SOC analyst

Real-time guardrails against prompt attacks

Generative AI for security operations

Automated AI security fixes for code

Security platform for AI/ML systems

Fast static analysis with AI triage

AI security analyst and threat hunting

Developer security with AI (incl. DeepCode)

AI-assisted supply-chain/dependency security

AI wearable that summarizes your day

Always-listening AI companion pendant

AI voice recorders that transcribe/summarize

Pocket AI companion device

AI smart glasses with camera and assistant

Open-source SDK that adds real-time multiplayer collaboration to AI applications.

Open-weight LLM rivaling top Anthropic and OpenAI models at a fraction of the cost.

Experimental P2P distributed OS for web and native apps built on Entity Core Protocol.

Claude MCP plugin that turns Jira tickets into merge-ready pull requests automatically.

Web-based GUI for the Pi coding agent, built in JetBrains' incubator.

Lets AI coding agents check an existing code library before generating solutions from scratch.

A free browser game where you battle head-to-head against your own computer.

Embeddable AI agent that lets your users build custom features on top of your product.

Real-time observability map tracking plans, tool calls, and file changes of AI coding agents.

Autonomous AI coding agent for Android Termux and desktop environments.

Serves markdown versions of web pages to AI agents via HTTP Accept headers.

Curated GitHub list of AI tools spanning IDEs, coding agents, CLI utilities, and more.

Verifies AI-generated code before coding agents mark it done.

Curated GitHub list of 200+ real AI products, apps, and bots built by Thai developers.

Scans Microsoft Foundry toolboxes and AI agents for governance, identity, and data-leakage risks.

MCP server that lets teams share AI memory, skills, and agents across projects.

Google's fast, lightweight multimodal Gemini model built for developers to integrate.

Open-source framework for running scenario tests on AI agents before deployment.

MCP gate that catches hallucinated APIs, fake test-pass claims, and slopsquatting packages.

Open-source, self-hostable AI coding assistant cloning Instinct's core features.

Open-source AI agents that autonomously test web applications for bugs and issues.

Syscall-level supervision and sandboxing for AI coding agents on Linux.

Coordination layer that lets multiple AI coding agents collaborate without conflicts.

MCP server that adds configurable, gated loops to AI coding agents.

Bridges AI agents so one model reasons and plans while another writes the code.

Open-source AI gateway for coding agents with model routing, fallbacks, and local model support.

Persistent, project-local memory for AI coding agents via MCP.

Free web search API for AI agents that helps abort hallucinations.

Open standard for structuring documentation so AI agents, not humans, can parse it.

Open-source project for running the latest iOS and macOS inside QEMU.

Open-source tool to define and enforce coding norms across multiple AI coding agents.

Open-source starter kit offering 198 modular skills for building Claude-powered AI agents.

Open toolkit for reproducible, standardized evaluation and benchmarking of TTS models.

Publish one OKF knowledge base that's readable by both humans and AI agents.

Semantic search MCP server that lets AI agents discover and query code packages.

Compresses prompts and repo context to cut token costs for Claude, Codex, and Cursor.

Open-source library enforcing per-argument authorization checks on AI agent tool calls.

Open-source library of reusable email components and layouts for developers.

A shared project brain where AI agents write context and people read it.

Access your computer's terminal, files, and browser from your phone.

Monitor, surface, and fix issues in agentic AI applications 24x7.

Practical, engineer-tested skills to make AI coding agents genuinely useful in real workflows.

Isolate AI coding agents safely in Podman containers with one bash script.

MC/DC test coverage checks that let AI coding agents verify code safely overnight.

Serverless reverse geocoding library for ESP32, RP2040, and browser environments.

Rust identity infrastructure providing authentication for both humans and AI agents.

Local-first search layer for fast code and text retrieval by humans and AI agents.

Identity-aware SSO and WireGuard gateway for securing self-hosted services and LLM access.

Coding agent constrained by a policy file living in your repository.

AI reverse engineering grounded on deterministic tools, with results verified against binaries.

Claude.ai API-compatible gateway with account pooling, streaming, and tool-use support.

Open-source multi-model agent runtime on Spring Boot with RAG, MCP tools, and code sandboxing.

Open-source, telemetry-free AI coding harness and editor for private development.

Open spec that scores the quality of MCP tool definitions for better LLM tool use.

An experimental AI agent that works directly in your terminal with real command line experience.

Shared procedural runbook memory for AI coding agents over MCP.

A registry that supplies context to AI coding agents to improve generated code.

Fast terminal assistant that suggests commands without using LLMs.

Open-source e-ink bike computer you can build, flash, and hack yourself.

AI runtime coding agent with a built-in 3D game engine for creating interactive games.

Open-source governance framework for controlling and coordinating multi-agent AI systems.

Retro-Mac styled ESP8266 desk device that mirrors your AI coding agent's live status.

Git-native persistent memory layer giving AI coding agents durable context across sessions.

Gives AI coding agents persistent memory of your codebase across sessions.

Lets AI coding agents control Chrome without stealing window focus.

Shadcn-style UI components for quickly building quiz funnels in web apps.

Sync and manage Claude.md and AGENTS.md files across machines and AI runtimes.

Review site for dev tools where the reviewers are autonomous AI agents.

iOS CI/CD pipeline designed to be driven by AI agents, no Mac required.

Code intelligence infrastructure giving frontier AI models deep codebase understanding.

Spatial canvas IDE for orchestrating AI coding agents across projects.

Essay arguing AI-driven coding marks the end of traditional software development.

Open source AI coding agent focused on efficiency.

Visual builder that generates FFmpeg commands for common video encoding workflows.

Hides encrypted messages inside ordinary-looking text using steganography.

Helps developers pick the right AI models, frameworks, and tools for their stack.

Fast, isolated execution sandboxes for safely running AI agent code.

Visualizes the token cost of AI coding agents reading whole files.

Lightweight framework that cuts insecure AI-generated code output by 72%.

Static hosting built so AI coding agents can deploy and manage sites end to end.

Open-source firewall that understands why your AI agent ran a given command.

Sign and verify AI instruction files with Sigstore to ensure integrity and provenance.

Open-source simulation infrastructure for testing AI voice agents.

TypeScript superset that makes LLM inference a native language feature.

SSH and Mosh terminal built for running Claude Code and AI coding agents remotely.

Live map visualizing your AI coding agents and their activity in real time.

Catalogs every CLI tool on your Mac with a JSON-friendly interface for AI agents.

Minimal open-source tool for hosting static websites with minimal setup and configuration.

Pipes real user feedback into your coding agent via MCP.

Open-source AI harness and starter kit for building AI-powered design apps.

Architecture change sets that keep AI coding agents aligned with your system design.

Research blog on SkySynth, an agent-based approach to building trustworthy specialized AI systems.

SimTower decompiled and rewritten as a playable game in your web browser.

A working C compiler built with AI assistance from Claude, showcased on GitHub.

Essay arguing developers should not design software tools specifically for AI agents.

PoC scanner and audit module for Windows Push Notifications use-after-free and race condition flaws.

Curated hub of resources for migrating legacy code to Rust and eliminating memory-safety bugs.

MCP server letting AI assistants create and edit UML diagrams in Astah Pro.

An IDE built for designing, testing, and researching agentic AI systems.

Open-source AI security terminal with vulnerability scanning and autonomous pentesting agents.

QA platform for testing and validating AI agents before deployment.

Lightweight, security-first AI agent runtime in pure Python with zero frameworks.

A minimalist self-mutating AI coding agent built as one file with stdlib only.

Self-hosted AI agent with WebUI, CLI, MCP tools and deterministic Record & Replay debugging.

AI-powered CLI that automates dependency upgrades and migrations using codemods.

Ready-made UI transition components designed for AI agent interfaces.

Open standard for capturing software intent to guide AI coding agents.

Spec-driven agentic platform that turns structured requirements into production-ready software.

Build and launch AI apps quickly with built-in security checks and guardrails.

Orchestrate multiple AI coding agents in parallel across different projects.

A design-quality-focused UI component library positioned as an alternative to Shadcn.

API that gives LLMs persistent state and memory across multiple providers.

Desktop tool for fast, locally indexed search across Kafka messages.

Generate and share AI-built apps with a single click.

Easily installed skill that generates visual change preview pages for code.

Framework for AI agents that write and run sandboxed Lua inside Elixir apps.

AI coding agent that runs in your terminal via an interactive TUI.

Self-hosted persistent memory for AI coding agents via MCP.

Self-hosted AI characters with persistent memory, shared worlds, and everyday-life tools.

Tiny open-source ledger that gives AI bots persistent, auditable memory.

Agent skill with multiple gates to detect and remove AI slop from generated code.

Proof-of-concept exploit chain for n8n unauthenticated file-read and expression-injection RCE vulnerabilities.

PoC module targeting AD CS misconfigurations for domain controller impersonation via PKINIT.

Vim-style terminal power tool for LLMs that slashes token usage for hyper-efficient workflows.

Testing and evaluations platform for MCP servers.

Open-source LangChain-like harness for building with Jev and other System 1 models.

Local-first Rust tool that adds an undo button for AI agents, no cloud required.

Define voice agents once with a single spec and deploy them to any stack.

Open evaluation framework for benchmarking AI agent memory Add/Search performance.

Evidence-based AI agent that diagnoses ROS 2 robots via tool-calling and safety gates.

VSCode extension that lets web AI tools manage local files via the MCP protocol.

Framework for building self-describing CLI tools that AI agents can discover and use.

The ultimate AI-powered video editing software.

Local code review tool built for agentic AI development workflows.

Open-source UML viewer for visualizing and managing Grok AI agents.

Fast, lightweight local-first harness for building and running AI agents.

Git host for AI agents with no account, token, or key required.

Local-first, zero-telemetry proxy daemon for routing and controlling AI tool traffic.

OAuth 2.1 authentication library for securing AI agent swarms.

Watch AI agents compete and battle against each other in a live arena.

Visual JavaScript app builder for creating and deploying custom web apps without heavy coding.

Previews what an AI agent's command would destroy before it executes.

A catalog of open-source AI tools for agents, MCP, skills, and RAG.

eBPF exporter that streams AI-agent activity metrics into Grafana.

Desktop tool to build, evaluate, and monitor AI agents.

Open-source Rust LLM gateway with a tamper-evident audit ledger.

Semantic diff tool that helps review code changes intelligently.

Open protocol for identity and accountability of AI agents.

Open-source Gaussian splat talking-head avatars for real-time AI agents.

macOS Quick Look extension adding syntax highlighting for code files in Finder.

Isolated local sandbox environments for running AI coding agents safely.

Sandboxed, observable AI coding agents for safe autonomous code execution.

Secure, private runtime environment for safely executing autonomous AI agents.

Open-source gateway that secures, governs, and observes AI agents' MCP tool calls and LLM traffic.

107 browser-only developer utilities that run locally with zero network calls.

Brings AI assistance into sensitive terminal environments in a classic Unix-style way.

Three AI agents write, test, and return your project as a working ZIP file.

UI for answering AI agent questions asynchronously to keep workflows unblocked.

Organizes and lets you review HTML pages generated by AI tools in one place.

Open-source npm package to track and measure how your AI coding agent performs.

Stateful Linux microVMs providing isolated, persistent environments for AI coding agents.

Benchmark coding agents against real-world software engineering tasks.

Open-source project enabling instinct-like decision-making behaviors for AI agents.

Analyzes a repo's Git history to produce acquisition-ready technical due diligence reports.

Lets AI agents deploy full-stack apps from prompts or workflows.

AI-powered tool for reverse engineering and analyzing app behavior.

Hardware keypad showing which AI coding agent needs your attention at a glance.

Run lightweight virtual machines on Apple Silicon, ideal sandboxed environments for AI agents.

Desktop companion that monitors AI coding agents, catches faked test passes, and verifies real results.

Local OpenAI-compatible gateway giving coding tools access to free AI models.

Manage and orchestrate multiple AI coding-agent sessions from one dashboard.

TypeScript framework for building and running AI agents.

AI agents fill typed holes in code while an SMT solver rejects incorrect fills.

Evals and observability platform for WebMCP servers and AI agent integrations.

Visualize, diff, and roll back changes in your AI agent's memory.

Fast shell command to jump between recent AI coding sessions from Oh My Zsh or Bash.

Controls which files AI coding agents can read, blocking sensitive file access.

Coding agent that cuts repeated-work API calls by up to 78%.

Define specifications and review what AI coding agents build against them.

Open standard for UI components usable by people, developers, and AI agents.

Analyzes how readable and usable your documentation is for AI agents.

Proxy that monitors and can roll back changes AI agents make to your database.

CLI that configures free AI models inside coding tools like Codex or VS Code with one command.

Harness that generates custom AI agents with their own tools, views, skills, and UIs.

Spring Boot starters that turn GraphQL operations into MCP tools and Spring AI tools.

Local on-device guardrail that judges every AI coding agent tool call to allow, review, or block.

A 1-bit quantized 1.7B parameter LLM that outperforms Bonsai-1.7B.

Agent skills and method for AI coding agents that build software for someone else.

Open-source agentic testing framework for web, iOS and Android apps.

Google's Gemini 4 Argon model, available for developers to build with via API.

Learns reusable skills from your coding agent corrections to improve future sessions.

Open source installer and skill packs to add tools to AI apps in one click.

Open-source benchmark for evaluating coding agents on real app deployment tasks.

Fast open-weight coding model by JetBrains, built to power coding agents.