IDE & editor agents
IDE or editor extension agent
twinnydotdev/twinny Twinny (twinnydotdev/twinny) is the clear leader—particularly if you want a privacy‑first, VS Code–centric assistant that can run on models you control; before choosing, decide whether your priority is local/BYOK privacy or specialized capabilities like auditability, native editor plugins, or multi‑agent UIs.
Which one matches your setup?
Answer any of the questions — the shortlist updates as you go. Recommendations come from the capability passports below, nothing else.
Comparison matrix
Axes are extracted from each project's docs by our review pipeline; the maturity score is computed from stars, growth and commit activity — not an opinion. Click a column to sort.
| Runs in | Models | Context | Cost to run | |||||
|---|---|---|---|---|---|---|---|---|
⭐ 1.9k +62/7d | IDE | OpenAIAnthropicFixed provider | Multi-file workspace | Approve/reject review | ●●●●● | Cloud-connected desktop | ●●●●● | Free desktop app; external model API costs may apply |
⭐ 3.7k +8/7d | IDECLIPR botWeb app | BYOKOpenAIAnthropicGeminiOllama / local | Whole-repo analysis | Rerank & review | ●●●●● | Fully local | ●●●●● | Free; team gateway paid after 5 devs |
⭐ 2.6k | CLIIDE | BYOK | Whole-repo analysis | Reviewable outputs | ●●●●● | Self-hostable | ●●●●● | Uses your API key (provider billing applies) |
⭐ 2.6k | CLIIDE | BYOK | Whole-repo analysis | Review controls | ●●●●● | Self-hostable | ●●●●● | Your API keys; free desktop app |
⭐ 169 +21/7d | Web appIDE | BYOKOpenAIAnthropic | Whole-repo index | Evidence & gating | ●●●●● | Hosted demo + local | ●●●●● | Free demo; may use external model APIs / your API key |
⭐ 941 +32/7d | IDECLICoding-agent plugin | Anthropic | Project-wide | Configurable tools & perms | ●●●●● | Anthropic cloud via CLI | ●●●●● | Requires Anthropic (Claude) account / subscription |
⭐ 870 +1/7d | IDE | BYOKOpenAIOllama / localFixed provider | Whole-repo analysis | Approval & review | ●●●●● | Self-hostable | ●●●●● | Free local preview; provider charges apply for BYOK/hosted models |
⭐ 656 +60/7d | IDE | AnthropicBYOK | Project workspace | None mentioned | ●●●●● | Cloud APIs (local keys) | ●●●●● | Requires cloud model API keys |
⭐ 292 | IDECLI | OpenAIAnthropicGeminiFixed provider | Diff + related files | Basic config filters | ●●●●● | Cloud APIs (user key) | ●●●●● | Varies by backend (cloud API or local CLI) |
⭐ 197 +1/7d | IDE | OpenAIAnthropic | Related files | None mentioned | ●●●●● | Third-party cloud | ●●●●● | Uses external cloud AI providers (may require API keys) |
⭐ 197 +1/7d | IDE | OpenAIAnthropicFixed provider | Open project files | None mentioned | ●●●●● | Cloud APIs | ●●●●● | Free to download; external model API fees may apply |
⭐ 853 | IDECoding-agent pluginWeb app | BYOK | Session-only | None | ●●●●● | Agent cloud dependent | ●●●●● | Uses your agent's tool-call quota |
Capability profiles
Six axes, 0–5 each. The shape tells you the strategy: a wide hexagon is a generalist, a spike is a specialist. Showing the 8 most established — the rest are in the full catalog.
A visual WYSIWYG workspace that runs multiple coding agents in parallel and lets you review and approve their edits as red/green diffs while managing sessions and tasks.
A privacy-first VS Code assistant that runs on models you control (local, peer devices, or a team gateway) with no telemetry or sign-in.
Provides persistent, replayable agent trajectories, cross-session memory, and AI-blame for team-level review and accountability that typical IDE agents do not offer.
Provides replayable, inspectable agent trajectories, cross-session memory, and AI-blame for team review and accountability—capabilities not offered by typical stateless AI IDEs.
Builds a compiler-style, tree-sitter whole-repo index and answers via structured queries and runtime traces instead of dumping the repo into an LLM context.
Native Unreal Engine 5.7 editor integration that exposes a dynamic UE5.7 context loader and a 20+ tool MCP bridge to Claude Code for direct Blueprint, asset, and level manipulation.
A durable, local-first agentic IDE that preserves task workspace and history across sessions while exposing explicit approval and review surfaces for sensitive actions.
Runs a full Ubuntu userspace and integrated AI agents directly on ARM64 Android devices, enabling on-device builds, terminal access, and live web previews in a single mobile app.
All repositories (18)
Open-source AI coding assistant for VS Code. Coding agent, code completion, chat, edits and reviews with local…
System of record for how agents build software: Built-in rust harness & 20+ CLIs.
System of record for how agents build software: Built-in rust harness & 20+ CLIs.
Nimbalyst - The open-source visual workspace for Claude Code, Codex, and OpenCode. Run multiple coding agents…
Claude Code CLI integration for Unreal Engine 5.7 - Get AI coding assistance with built-in UE5.7 documentation…
Local-first, zero-trust agentic IDE for verifiable autonomous software development.
10x-Tool-Calls is a lightweight rules file designed to help you maximize the value of your monthly AI coding t…
Claude Code on Android: AI-powered mobile coding IDE for Android — chat with a coding agent, run Linux comman…
IDE, A native-feeling AI coding workspace that blends chat, planning, agent execution, and project navigation…
IDE, A native-feeling AI coding workspace that blends chat, planning, agent execution, and project navigation…
Unified Emacs interface supporting OpenAI Codex, Antigravity CLI, Claude Code, Pi agent, Muse Code, and more
OpenSpec Plus — Agentic skills that enhance OpenSpec's Spec-Driven Development through better discovery, requi…
Most coding agents grep and glob raw source into LLM's context window and hope the model is smart. Benzi compi…
A vault-specific agent equipped with agentic capacity, fast search, flexible commands, vault management, and t…
Hosted alternatives
If running your own reviewer is more ops than you want, these managed services cover the same job.
The managed AI-first editor: the agent, model access and indexing come on a subscription, so there is nothing to self-host or wire to a provider.
Try Cursor →Codeium's agentic editor with the coding agent built in and hosted, an alternative to bolting an open-source agent onto your own editor setup.
Try Windsurf →Managed in-editor assistance and agent mode across the major IDEs on a subscription, run against GitHub's models rather than a harness you install and configure.
Try GitHub Copilot →