38 AI Coding Tools Quick-Reference (2026)
Source: all 38 tools in the ai-code category of 9bests.com AI tool library ｜ Generated: 2026-07-14 ｜ Ratings are 9bests editorial scores, independent of vendor payments.
Subscribe to the 9bests weekly — get the full list free
Hand-picked AI tool reviews and updates every week. Subscribe to receive this full list + 7 more quick-reference sheets (writing / image / video / audio / chat models / data / API cost).
Subscribe free & get it →Independent reviews — ratings aren't influenced by vendor payments · double opt-in · unsubscribe anytime
1. 30-second decision guide (grouped by objective attributes)
① Zero-cost / self-hostable OSS (16)
Cate 4.7 ｜ Sync 4.6 ｜ OpenYabby 4.5 ｜ Polis 4.4 ｜ git-lrc 4.4 ｜ Drydock 4.4 ｜ Agent Joe 4.2 ｜ smolfs 4.2 ｜ Selvedge 4.2 ｜ cap'n hook 4.2 ｜ Leakproof 4.1 ｜ Crawlie 4.1 ｜ Lowfat 4 ｜ Proctor 4 ｜ Lathe 3.9 ｜ agent-run 3.8
② Has a free tier to try (16)
Cursor 4.8 ｜ v0 by Vercel 4.5 ｜ Windsurf (Codeium) 4.4 ｜ Lovable 4.4 ｜ Lore 4.4 ｜ Replit Agent 4.3 ｜ Bolt.new 4.3 ｜ Command Center 4.3 ｜ Freebuff 4.2 ｜ Rayline 4.1 ｜ Righthand 4.1 ｜ MCPlexer 3.9 ｜ Prototyper 3.9 ｜ Orchestrator 3.8 ｜ Socrates 3.6 ｜ DesktopMCP 3.25
③ Paid / team-tier (3)
Lobsteady 4.7 ｜ GitHub Copilot 4.6 ｜ Claude Code 4.5
④ Price undisclosed / quote (3)
Ito.ai, Infraas.ai, Toolnexus
Prices are per each tool's official site and may change at any time; this list is for selection reference only, not purchasing advice.
2. Full comparison table (sorted by rating)
| # | Tool | Rating | Price | One-line intro | Note |
|---|---|---|---|---|---|
| 1 | Cursor | 4.8 | Free / $20/mo | AI-centric code editor built on VS Code | Based on a VS Code fork — some reservations |
| 2 | Lobsteady | 4.7 | Paid (from $20/mo) | Runs autonomous coding agents on Telegram, Discord and Slack at a flat $20/mo | Learning curve |
| 3 | Cate | 4.7 | Free (Open Source) | Open-source canvas IDE giving a visual workspace for agentic coding workflows | Learning curve |
| 4 | GitHub Copilot | 4.6 | $10–39/mo | AI pair-programming assistant by GitHub and OpenAI | Requires subscription |
| 5 | Sync | 4.6 | Free (Open Source, Alpha) | Quality layer for AI agents that pins code changes to auto-flag stale context | Learning curve |
| 6 | v0 by Vercel | 4.5 | Free / $20/mo | AI UI component generator for React/Next.js | Frontend only |
| 7 | Claude Code | 4.5 | Included in Claude Pro | Anthropic's terminal-based AI coding agent | CLI only |
| 8 | OpenYabby | 4.5 | Free (Open Source) | Voice-controlled multi-agent orchestrator built on Realtime API + CLI runner | Learning curve |
| 9 | Windsurf (Codeium) | 4.4 | Free / $15/mo | Free AI code completion and chat assistant | Weaker than Copilot |
| 10 | Lovable | 4.4 | Free / $20/mo | AI full-stack app builder that generates production-ready web apps from natural language | Complex apps still need manual polish |
| 11 | Lore | 4.4 | Freemium | LLM proxy for coding agents that manages shared cross-session context & memory | Learning curve |
| 12 | Polis | 4.4 | Free (Open Source, MIT) | Multi-agent control plane coordinating Claude Code, Codex, etc. via contracts | Learning curve |
| 13 | Ito.ai | 4.4 | Not disclosed | Code-review tool that actually runs code, finds more bugs at lower false-positive rates | Learning curve |
| 14 | git-lrc | 4.4 | Free (Open Source) | Micro AI code reviewer triggered on git commit; opens a review UI across 10+ repos | Learning curve |
| 15 | Drydock | 4.4 | Free (Open Source) | Hardware-isolated VM sandbox for macOS autonomous coding agents | Learning curve |
| 16 | Replit Agent | 4.3 | Free / $25/mo | AI cloud IDE that can build complete apps | Performance ceiling |
| 17 | Bolt.new | 4.3 | Free / $20/mo | AI tool for building full-stack apps in the browser | Complex apps still need editing |
| 18 | Command Center | 4.3 | Freemium | AI coding environment that reviews multi-file changes via interactive walkthroughs | Learning curve |
| 19 | Agent Joe | 4.2 | Free (Open Source, GPL-3.0) | Open-source TUI Rust coding assistant with no shell access, safer | Learning curve |
| 20 | Freebuff | 4.2 | Free (ad-supported) | Ad-sponsored free CLI coding agent, alternative to Claude Code / Codex / Cursor | Learning curve |
| 21 | smolfs | 4.2 | Free (Open Source, Apache 2.0) | Persistent cross-session workspace volume for AI coding agents with local SQLite | New project, small community |
| 22 | Selvedge | 4.2 | Free (Open Source, MIT) | Local MCP server giving AI coding agents context on 'why code works' | Requires the agent to call its MCP tools |
| 23 | cap'n hook | 4.2 | Free (Open Source, MIT) | Lightweight, local-first memory tool for coding agents | First run downloads 300MB–2GB embedding model |
| 24 | Rayline | 4.1 | Free (pay per API use) | Routing layer for coding agents; routes each sub-agent call to the best model by quality/speed/cost | Learning curve |
| 25 | Righthand | 4.1 | Freemium (7-day free trial) | Autonomous AI employee with skills, goals and CLI that works in your tools | Learning curve |
| 26 | Leakproof | 4.1 | Free (Open Source, Apache 2.0) | Prevents secret leakage for AI assistants like Claude Code / Cursor / aider | Learning curve |
| 27 | Crawlie | 4.1 | Free (Open Source, MIT) | Fast, free, open-source technical SEO + GEO crawler for humans and agents | Learning curve |
| 28 | Lowfat | 4 | Free (Open Source) | Pluggable CLI filter that trims command output noise, saving agents up to 91.8% tokens | Filtering may drop critical errors |
| 29 | Proctor | 4 | Free (Open Source) | Kernel-level sandboxing + ed25519 signatures to stop AI agents from unsafe actions | Niche (benchmarking) |
| 30 | Toolnexus | 4 | Unknown | Provider-agnostic toolkit: dynamic MCP server generation | ⚠️ TBD |
| 31 | Lathe | 3.9 | Free (Open Source) | CLI that uses LLM skills to generate hands-on multi-chapter technical tutorials with a local Web UI | LLM-generated tutorials less reliable than human-written |
| 32 | MCPlexer | 3.9 | Free | Unified MCP OS-layer with directory-level routing, persistent workers, task management | New project, fast iteration |
| 33 | Prototyper | 3.9 | Free | Visual workspace for AI agents and teams | ⚠️ TBD |
| 34 | Orchestrator | 3.8 | Free (BSL 1.1, source-available) | Local CLI + agent skill to run and supervise multiple agents from one interface | BSL license restricts commercial hosting |
| 35 | agent-run | 3.8 | Free (Open Source, GPL-3.0) | <1MB standalone binary that runs agents inside Bubblewrap | Linux only (needs bwrap & user namespaces) |
| 36 | Infraas.ai | 3.75 | Unknown | AI-powered cross-repo bulk code-change platform: NL description → auto PR + CI verify | New product, low community activity (GitHub 2 stars) |
| 37 | Socrates | 3.6 | Free | A multi-agent protocol from Hexo Labs | ⚠️ TBD |
| 38 | DesktopMCP | 3.25 | Free | MCP server giving AI models 144 Linux desktop interaction tools with vision | Linux only, no macOS/Windows |
Prototyper, Socrates, Toolnexus have TBD in the "Note" column (under editorial review) — not flawless; intros are taken from each tool's official description.
3. Four questions to narrow your choice
Budget
0 → see ① OSS; can pay $10–20/mo → see ②③.
Deployment
Data stays local / need self-host → prefer OSS self-hostable.
Scenario
Code completion · monorepo refactor · generate UI · multi-agent orchestration — strengths differ; see the 'Intro' column.
Lock-in risk
Bound to a single model / editor / platform? See the 'Note' column.
Want the full version?
This list is updated quarterly (new tools, price changes, rating reviews). Subscribe to get: the updated list + 7 more same-style quick-reference sheets for other segments (writing / image / video / audio / chat models / data / API cost).
Subscribe to get the full list →9bests.com ｜ Independent AI tool reviews — ratings aren't influenced by vendor payments.