JetBrains AI Assistant Alternatives for IntelliJ, PyCharm and More

If you work in IntelliJ IDEA, PyCharm, WebStorm or another JetBrains IDE, alternatives to JetBrains AI Assistant include GitHub Copilot, Claude Code, Codex, Cline, Kilo Code, OpenCode, Tabby and ProxyAI. They differ in licence, models, where they run and how they ask before acting. Safeguard adds a JetBrains extension and a CLI that knows your vulnerabilities.

Switching to Safeguard does not cost you the plan you already have. Bring the Claude, ChatGPT or Codex plan you already pay for and use it inside the Safeguard CLI. Safeguard adds no extra charge for that usage; your plan's own usage limits and terms still apply. You can also connect Claude or ChatGPT to the Safeguard MCP server and use Safeguard from the chat you already work in.

What JetBrains AI Assistant includes today

JetBrains describes AI Assistant as "a collection of AI-powered features and coding agents integrated into JetBrains IDEs". That covers AI Chat, code generation and completion, explanations and refactoring help, and routine work such as tests, documentation, commit messages and pull request summaries. The docs list CLion, DataGrip, DataSpell, GoLand, IntelliJ IDEA, PhpStorm, PyCharm, Rider, RubyMine, RustRover and WebStorm.

Agents inside AI Chat. The agents page lists four built-in agents, each with its own modes: Junie (Brave mode, Debug mode), Claude Agent (Auto, Default, Accept Edits, Plan, Don't Ask, Bypass), Codex (Read-only, Agent, Agent full access) and GitHub Copilot (Agent, Plan, Autopilot). You approve, deny or let an agent run automatically depending on its mode, then keep or roll back the changes it proposes.

External agents through ACP. The Agent Client Protocol lets any agent that implements it join AI Chat. You install one from the ACP registry inside the IDE or add your own in ~/.jetbrains/acp.json. JetBrains notes that ACP agents work without a JetBrains AI subscription but are not supported in WSL.

Junie. Junie is JetBrains' own coding agent. It runs inside JetBrains IDEs, from the terminal after a curl install, as a GitHub Action, and from GitLab issues or merge requests. Its site lists a free Junie Lite model, bring-your-own-key use and the AI Pro and AI Ultimate plans. The Junie CLI repository's README says "All rights reserved" and that use is subject to the JetBrains AI Service Terms.

Models and your own keys. You can add Anthropic, Google, OpenAI or any OpenAI-compatible endpoint with your own key, and run local models through Ollama or LM Studio. Two limits are worth knowing: agents cannot use local models, and AI Assistant cannot call MCP tools while you use a local model.

Plans. The licensing page lists AI Free, AI Pro, AI Ultimate and AI Enterprise, plus a trial. Quotas are counted in AI credits that renew every 30 days. AI Pro comes with the All Products Pack and dotUltimate, though not when your IDE licence comes from a License Server, and AI Free is not available in Android Studio or the Community editions of PyCharm and IntelliJ IDEA.

That is a broad set of options inside one IDE. People look elsewhere when they want a different licence, a specific vendor's agent, a self-hosted server or a tool their team already uses outside JetBrains.

Alternatives for JetBrains IDEs compared

Every tool below has a JetBrains route we confirmed in its own docs. For the open-source ones, we confirmed the licence on the repository page. "Where code runs" means where the tool executes; prompts and the files an agent reads still go to your model provider unless you use a local model.

GitHub Copilot

GitHub lists JetBrains IDEs among Copilot's supported editors, and its install guide names more than a dozen of them, from IntelliJ IDEA and PyCharm to Rider and RustRover. You install the plugin from the Marketplace, restart, then sign in with a device code from Tools, GitHub Copilot, Login to GitHub. Know before adopting: you need Copilot Free or a paid Copilot plan, and JetBrains also lists Copilot as a built-in agent in AI Chat, so check which of the two routes your team means.

Claude Code

Claude Code's JetBrains plugin runs the claude command in the IDE's terminal and connects to it, adding diffs in the IDE's diff viewer, selection sharing and access to the IDE's inspection diagnostics. It does not bundle the CLI, so you install both, and it works with a paid Claude plan or a Console account. Know before adopting: Anthropic's docs warn that in acceptEdits mode, Claude Code "may be able to modify IDE configuration files that can be automatically executed by your IDE", and suggest Manual mode for edits.

Codex

OpenAI's IDE docs say VS Code and compatible editors use the Codex extension, while "Xcode and JetBrains IDEs provide their own integrations". In JetBrains, you open AI Chat and select Codex. The Codex CLI itself is Apache 2.0 and signs in with a ChatGPT plan or an API key. Know before adopting: JetBrains lists three modes for Codex in AI Chat, Read-only, Agent and Agent full access, so pick one before the first task.

Cline

Cline installs from the Marketplace, opens from View, Tool Windows, Cline, and lets you choose the Cline provider, a ClinePass subscription or your own provider key. Its repository is Apache 2.0, but the README says "Currently we are not open-sourcing JetBrains plugins". In Act mode, every file edit and terminal command needs your approval unless you turn on auto-approve. Know before adopting: if open source is the reason you are switching, the JetBrains client is the part that is not.

Kilo Code

Kilo Code is an MIT-licensed agent with a VS Code extension, a CLI, a Cloud Agent and a JetBrains plugin. Its docs say v7 for JetBrains "is officially available", supports IntelliJ IDEA, WebStorm, PyCharm, PhpStorm, GoLand, Rider, CLion, RubyMine and DataGrip, and needs no Node.js. Know before adopting: the right-click Auto-Approve toggle applies to every Kilo session in the IDE, not only the current one, and a "Bundled Core" build exists for locked-down networks.

OpenCode

OpenCode is an MIT-licensed terminal agent that supports the Agent Client Protocol, so it can run inside JetBrains AI Chat. You add it to acp.json with the absolute path to the binary and select it in the agent picker. Its providers page lists more than 75 providers and local models. Know before adopting: its permission docs say most permissions default to allow. More in our OpenCode alternatives guide.

Tabby

Tabby is a "self-hosted AI coding assistant" that runs as your own server, with plugins for JetBrains, VS Code and Vim. The JetBrains plugin works with IntelliJ Platform IDEs from build 2023.1 and needs Node.js 18 or later. Its LICENSE file applies Apache 2.0 to everything except the ee/ directory, which has its own licence. Know before adopting: you run and update the server, and the newest server release on the README is v0.30.

ProxyAI

ProxyAI is an Apache 2.0 plugin built for JetBrains IDEs. It connects to hosted models from OpenAI, Anthropic, Azure, Mistral and others, or to a self-hosted model for what the README calls "a full offline experience". Know before adopting: the repository page showed no archived notice but no release date either, so check the Marketplace listing for the version that matches your IDE.

A note on Continue

Continue still has a JetBrains plugin, but its README now recommends the Continue CLI instead, and says the repository "is no longer actively maintained and is read-only for all users". Plan for no further updates if you rely on it.

Setting it up in your IDE, step by step

These steps assume IntelliJ IDEA, but the menus match across JetBrains IDEs. Try one tool on a branch before you roll it out.

Step 1: Branch and baseline. Open the IDE terminal in your project root and take a dependency baseline, so you can compare after the agent works:

git switch -c agent/try-new-assistant
curl -fsSL https://cli.safeguard.sh/install | bash
safeguard scan --dir .

Results land as JSON files under ./output, including vuln.json and license.json.

Step 2: Install the agent you want to try.

For Claude Code, install the CLI, then the plugin, then run it from the IDE terminal in Manual mode:

curl -fsSL https://claude.ai/install.sh | bash
claude --permission-mode manual

If you start claude in an external terminal, type /ide to connect it to the running IDE.

For OpenCode over ACP, install it, then add it to ~/.jetbrains/acp.json:

{
  "agent_servers": {
    "OpenCode": {
      "command": "/absolute/path/bin/opencode",
      "args": ["acp"]
    }
  }
}

Restart the IDE if it does not appear, then pick OpenCode in the AI Chat agent selector.

For GitHub Copilot, Cline, Kilo Code, Tabby or ProxyAI, open Settings, Plugins, Marketplace, search for the name, install and restart. Copilot signs in with a device code. Tabby asks for your server endpoint in its status bar settings, with http://localhost:8080 as the default.

Step 3: Set approvals before the first task. In Kilo Code, open Settings, Tools, Kilo Code, Auto-Approve and set each tool to Allow, Ask or Deny. In Claude Code, Shift+Tab cycles modes in the IDE terminal. In AI Chat, choose the agent's mode from its selector.

Step 4: Run one real task. Ask for something small with tests, such as validation on one endpoint. Review the change in the IDE's diff viewer and run the tests yourself.

Step 5: Check the result before you merge.

git diff main --stat
safeguard scan --dir .
safeguard exec "summarise the critical findings in this repository"

A second scan shows whether the change pulled in a dependency with a known vulnerability. Our guide to checking coding agent changes covers this in more depth.

Common mistakes in JetBrains setups

Letting an agent edit IDE configuration. Anthropic's JetBrains docs warn that acceptEdits and auto mode approve edits in your working directory without asking, which can reach IDE configuration files the IDE runs. Use Manual mode on unfamiliar repositories.

Expecting local models to drive agents. JetBrains says agents cannot use local models, and MCP tools are unavailable with them. Use local models for chat and completion, and a hosted model for agent work.

Trying ACP agents inside WSL. JetBrains says ACP agents are not supported in WSL. Run them on the host, or use a tool with its own plugin.

Installing on the wrong side of Remote Development. For Claude Code, the plugin goes on the remote host through Settings, Plugin (Host), not on your local client.

Opening the IDE's MCP port to the network. Claude Code's plugin has an option to accept connections from all network interfaces. Its docs say the traffic and token then cross the network unencrypted, so leave it off unless loopback cannot work.

Leaving a global auto-approve switched on. Kilo's toggle covers every session in the IDE. Check it when you open a new project.

Treating approval as review. An approved command can still add a vulnerable package. Scan the change before merging, whichever agent wrote it.

Do the work properly with the Safeguard JetBrains extension and CLI

Every tool above can write code. Safeguard writes code too, and it starts from your Safeguard data: the vulnerabilities in your dependencies and your scan results. The CLI scans code, container images and SBOMs on your own machine, sends only the results to the platform, and includes an assistant that can fix what it finds.

In the IDE

The download page lists a JetBrains extension, version 5.0.1, for "IntelliJ IDEA, WebStorm, PyCharm, GoLand & more", and describes the IDE extensions as "inline security scanning, guardrail enforcement, and one-click remediation inside your editor". Install it from the Install Extension button on that page. The same page lists extensions for VS Code and for Open VSX editors such as Cursor and Windsurf.

Keep the Claude, ChatGPT or Codex plan you pay for

Bring the Claude, ChatGPT or Codex plan you already pay for and use it inside the Safeguard CLI. Safeguard adds no extra charge for that usage; your plan's own usage limits and terms still apply. To connect a model, see the CLI docs or run /model in a session. You can also connect Claude or ChatGPT to the Safeguard MCP server and ask about your findings from that chat.

Model choice works on the platform too. This demo shows choosing the model for each conversation. Zero z20 answers with no language model at all, turning the question into a platform query. Griffin, Eagle and Lion are Safeguard's models, a tenant admin chooses which model families people may use, and you can bring your own model on your own provider account and route jobs such as chat or summaries to it:

Click through it at your own pace, or open it full screen.

Install and first session in the IDE terminal

The CLI is a single static binary with no Node, Python or JVM needed, for macOS, Linux and Windows, on x86_64 or arm64, with Windows on x86_64 only. Run it from the JetBrains terminal so it sees the same project folder as your editor:

# macOS and Linux
curl -fsSL https://cli.safeguard.sh/install | bash

# Windows (PowerShell)
irm https://cli.safeguard.sh/install.ps1 | iex

safeguard

Inside the session, /login signs in through your browser and /scan . scans the current folder. @ attaches a file from the folder you started in, and /resume picks up the last session in that folder.

Writing and fixing code

The assistant has tools for local files and for your Safeguard data and picks whichever the question needs. Ask it to "fix the lodash CVE in this repo" and it works from your findings, then writes files, edits code and runs commands, with your approval. The changes land in your project, so you review them in the IDE's own diff and history views. It also works as a general terminal assistant; the docs give "what's the latest file in ~/Downloads" as an example. Read more in the Safeguard CLI as an AI assistant on your machine.

This demo attaches a computer running the Safeguard CLI and a folder, asks for a fix, reviews the diff and approves the command before it runs. The demo notes that computer work from Search is available and expanding:

Click through it at your own pace, or open it full screen.

Approvals

Tools that change things ask before they act. /permissions sets how often: Ask prompts before every action, Auto approves known actions and asks for anything new (the default), and Yolo approves everything. On a repository you do not know well, start in Ask.

CI with safeguard exec

safeguard scan --dir .
safeguard scan --image your-image:tag
safeguard exec "summarise the critical findings in this repository"

safeguard exec runs one non-interactive assistant turn for pipelines, and safeguard validate checks CycloneDX, SPDX and VEX documents. Policies and gates are evaluated in the platform, not by the CLI. When the local runner handles platform scan jobs, your source code never leaves your machine; only SBOMs and findings are uploaded. The platform adds reachability analysis and autonomous remediation with fix pull requests, off by default.

Beyond the IDE

Start in the terminal, continue in the desktop app for Windows, macOS and Linux, and check findings on your phone with the iOS and Android apps. On Linux, the download page notes that the desktop AppImage needs the Safeguard CLI on your PATH. With nothing to install, the cloud sandbox gives you an isolated tenant against a seeded demo repo, free for seven days with no card.

FAQ

Can I use other AI agents inside JetBrains AI Chat?

Yes. JetBrains lists Junie, Claude Agent, Codex and GitHub Copilot as built-in agents, and any agent that supports ACP, such as OpenCode, can be added from the registry or through acp.json.

Do I need a JetBrains AI subscription to use these alternatives?

Not for all of them. JetBrains says ACP agents work without one, and Marketplace plugins such as Copilot, Cline, Kilo Code, Tabby and ProxyAI use their own accounts or keys.

Which alternatives are open source?

From the repository pages we read: Kilo Code and OpenCode under MIT, and Tabby and ProxyAI under Apache 2.0, with Tabby's ee/ directory under a separate licence. Cline's repository is Apache 2.0, but its JetBrains plugin is not open source.

Can I run a JetBrains assistant fully on my own hardware?

Tabby runs as a server you host, and ProxyAI and AI Assistant can use self-hosted or local models. JetBrains notes that its agents cannot use local models.

If I switch to Safeguard, do I lose my Claude, ChatGPT or Codex plan?

No. Bring the Claude, ChatGPT or Codex plan you already pay for and use it inside the Safeguard CLI. Safeguard adds no extra charge for that usage; your plan's own usage limits and terms still apply. You can also connect Claude or ChatGPT to the Safeguard MCP server and use Safeguard from that chat.

Does Safeguard replace JetBrains AI Assistant?

It is not a feature-for-feature replacement for AI Assistant or any tool here. It writes and fixes code with an assistant that knows your findings, and it scans what any agent changed. Many teams keep their current assistant and add Safeguard before merge.

Which JetBrains IDEs does the Safeguard extension support?

The download page lists the JetBrains extension for "IntelliJ IDEA, WebStorm, PyCharm, GoLand & more". The CLI runs in any IDE terminal on macOS, Linux and Windows.

Try it yourself

This walkthrough sets up the runner on your own machine: pick the platform, start it, approve it with its device code, then use the same CLI to scan from the terminal.

Click through it at your own pace, or open it full screen.

Switch without giving anything up. Download Safeguard for your JetBrains IDE and terminal, connect the Claude, ChatGPT or Codex plan you already have, and read the CLI docs. Weighing Claude Code specifically? See our Claude Code alternatives.

Quick check

Sources

• JetBrains: About AI Assistant
• JetBrains: Agents in AI Assistant
• JetBrains: Agent Client Protocol
• JetBrains: Custom and local models
• JetBrains: Licensing and subscriptions
• Junie
• Junie repository
• GitHub Docs: Where to use GitHub Copilot
• GitHub Docs: Install the Copilot extension
• Claude Code docs: JetBrains IDEs
• Claude Code docs: Overview
• Claude Code docs: Permission modes
• Codex docs: IDE extension
• OpenAI Codex repository
• Cline repository
• Cline docs: Installing in JetBrains
• Kilo Code repository
• Kilo Code licence
• Kilo Code docs: JetBrains
• OpenCode repository
• OpenCode docs: ACP
• OpenCode docs: Providers
• OpenCode docs: Permissions
• Tabby repository
• Tabby licence
• Tabby docs: IntelliJ plugin
• ProxyAI repository
• Continue repository
• Safeguard download page
• Safeguard CLI docs
• Safeguard CLI docs: Interactive terminal

Last reviewed: 21 October 2026.