RunVSAgent
Run selected VS Code coding agents and extensions inside JetBrains IDEs.
- Source repo
- wecode-ai/RunVSAgent
- Stars
- ★ 717
- Last updated
- 11mo ago
- License
- Apache-2.0
- Primary language
- Kotlin
- FA score
- 27/100 · Major gaps
At a glance
- How it runs
- Works with
- Platform-specific
- Cost
- Free software; you pay for model usage
- Setup effort
- Medium · a few setup steps
- You'll need
- Typical use
- A Java engineer using IntelliJ IDEA who wants to use Roo Code without moving their project workflow to VS Code.
- Not a fit if
- Developers who do not use JetBrains IDEs or run versions below 2023.1
- Teams needing a standalone CLI or a hosted agent service
- Source review
- 27/100 · Major gaps 8 safety controls not found
What does this agent do, and when should you use it?
RunVSAgent is a cross-IDE integration tool for running VS Code-based coding agents and extensions in JetBrains products such as IntelliJ IDEA, WebStorm, and PyCharm. Its architecture combines a Kotlin JetBrains Plugin with a Node.js Extension Host, communicating through RPC over Unix Domain Sockets or Named Pipes. The Extension Host supplies a VSCode API Layer, Agent Manager, and WebView support for the listed agents: Roo Code, Cline, and Kilo Code. It is delivered as a JetBrains Marketplace plugin, a ZIP release, or a source build. It fits teams that want to keep a JetBrains-centered development environment while using these specified VS Code agent integrations.
The JetBrains Plugin provides IDE integration, UI Integration, and an Editor Bridge, then communicates by RPC with the Node.js Extension Host. The host includes main.ts, extensionManager.ts, rpcManager.ts, and webViewManager.ts, alongside the VSCode API Layer, Agent Manager, and WebView support. This bridge is intended to run VS Code coding agents within a JetBrains IDE; the README names Roo Code, Cline, and Kilo Code. Cline is specifically described as being able to create or edit files, execute commands, and use the browser with permission at every step; that description is evidence for Cline, not a blanket guarantee for every supported agent.
- A Java engineer using IntelliJ IDEA who wants to use Roo Code without moving their project workflow to VS Code.
- A JavaScript or TypeScript team working in WebStorm that needs to bring Cline into its existing IDE.
- A Python developer in PyCharm who wants to run Kilo Code from the same IDE.
- A team standardizing across PhpStorm, RubyMine, CLion, GoLand, DataGrip, Rider, or Android Studio while adopting VS Code-oriented coding agents.
- A contributor debugging the bridge locally by running npm run dev in extension_host and ./gradlew runIde in jetbrains_plugin.
How do you install or deploy this agent?
Recommended: in a compatible JetBrains IDE, open Settings/Preferences → Plugins, search Marketplace for “RunVSAgent,” select Install, and restart when prompted. Alternatively, download the ZIP from GitHub Releases, choose Settings/Preferences → Plugins → the gear menu → Install Plugin from Disk..., select the ZIP, and restart. Source builds require Node.js 18.0+, a JetBrains IDE 2023.1+, Git, and JDK 17+: run git clone https://github.com/your-org/RunVSAgent.git, enter RunVSAgent, then run ./scripts/setup.sh and ./scripts/build.sh. The built plugin is placed in jetbrains_plugin/build/distributions/. The supplied material does not document how to install, configure, or first launch Roo Code, Cline, or Kilo Code after the plugin is installed, nor does it state required credentials.
How do you use this agent?
After installation and IDE restart, the README states that Roo Code, Cline, and Kilo Code can run in JetBrains IDEs, but it does not provide a menu path, configuration keys, extension acquisition procedure, or a first copyable agent invocation. A first agent run therefore cannot be verified from the supplied material. For development, run npm install followed by npm run dev in extension_host, then run ./gradlew runIde in jetbrains_plugin.
What are this agent's strengths and limitations?
- Uses a purpose-built Kotlin JetBrains Plugin plus Node.js Extension Host architecture to bridge JetBrains IDEs and the VS Code extension ecosystem.
- Explicitly uses RPC over Unix Domain Sockets or Named Pipes for local inter-process communication.
- Documents support across a broad set of JetBrains products, including IntelliJ IDEA, WebStorm, PyCharm, CLion, GoLand, Rider, and Android Studio.
- Names Roo Code, Cline, and Kilo Code as supported agents and includes WebView support in the documented architecture.
- Its operating boundary depends on the JetBrains IntelliJ Platform and Node.js, rather than being a standalone CLI or hosted agent service.
- It requires JetBrains IDE 2023.1+; source builds additionally require Node.js 18+, JDK 17+, Git, npm/pnpm, Gradle, and shell scripts.
- The material does not document post-install configuration, how the agent extensions are obtained, a first invocation, or credential setup.
- A known-issues document is referenced but its contents are not supplied, so common failure modes, limitations, and recovery guidance cannot be assessed.
How does this agent compare with similar options?
Compared with running these coding agents directly in VS Code, RunVSAgent is positioned as a bridge that brings specified VS Code agents and extensions into JetBrains IDEs through a JetBrains Plugin, Node.js Extension Host, and VSCode API Layer. No feature or performance benchmark is provided.
Key facts side by side with the most closely related agents.
| Agent | Source review | Form / cost | Stars | Updated | Language | Full support on |
|---|---|---|---|---|---|---|
| RunVSAgent This agent | 27 · Major gaps | IDE extensionFree + model costs | ★ 717 | 11mo ago | Kotlin | — |
| Kilo Code — The Open-Source All-in-One Coding Agent | 55 · Major gaps | IDE extensionFreemium | ★ 27k | 1d ago | TypeScript | — |
| Cline | 54 · Major gaps | CLIFree + model costs | ★ 69k | today | TypeScript | OpenAI API · Claude API |
| Continue | 43 · Major gaps | CLIFree + model costs | ★ 36k | today | TypeScript | — |
How does FollowAgents rate this agent?
Why each dimension lost points
Evidence: The repository provides no documentation on permission model, user confirmation, data flow, sensitive data handling, dependency security, external effects, rollback, or source attribution. Deduction: All trust criteria lack evidence, making it impossible to assess security and least privilege.
Evidence: The README's architecture description is consistent with the project structure, but no error handling or failure messages are documented. Deduction: Self-consistency is partially met; dependency availability is only mentioned as Node.js 18+ and JetBrains IDE 2023.1+, but dependency completeness is not verified; failure messages have no evidence.
Evidence: The README clearly lists supported IDEs and agents, and provides installation and build instructions, but does not define capability boundaries or trigger conditions. Deduction: Audience and scenarios are clear, environment fit is good, but capability boundaries and trigger precision lack documentation.
Evidence: The README has a clear structure with sections for installation, build, project structure, known issues, and includes an Apache-2.0 license. Deduction: Information architecture and install notes are adequate, but naming stability, examples and FAQ, known limitations, versioning/changelog, and maintenance responsibility are only partially mentioned without detailed evidence.
Evidence: The README describes core features and architecture, but provides no actual output examples or user feedback. Deduction: Output usability lacks evidence; marginal value is described but not quantified; cost-benefit is not discussed.
Evidence: Claims in the README (e.g., supported IDEs) are partially consistent with the project structure, but no external validation or test results are provided. Deduction: Claim traceability is partially met, but cross-source corroboration is absent, and facts vs. inferences are not clearly separated.
- Not found in source: least-privilege scopingGrant only what the task needs: a dedicated account or read-only token, scoped to specific directories and repos.
- Not found in source: confirmation before actingTurn on (or add) a confirmation step before it acts, and try it in a sandbox or test environment before real data.
- Not found in source: data-flow disclosureWatch which external services it contacts (proxy or firewall logs) and keep sensitive data out until you know where it goes.
- Not found in source: sensitive-data handlingUse dedicated, low-privilege, revocable API keys — never production credentials — and keep secrets out of logs.
- Not found in source: dependency securityPin versions and run a dependency audit (npm audit, pip-audit) before installing; prefer running it in a container.
- Not found in source: disclosed external effectsEstablish which external systems it writes to, sends to or changes, and verify with test accounts or repos before production.
- Not found in source: rollback or recovery pathBack up first, or work on a git branch or snapshot, so its changes can be undone.
- Not found in source: verifiable attributionInstall from the official repo or registry and check the publisher and URL to avoid look-alike packages.
- The repository provides no security documentation, such as permission model, data flow, or sensitive data handling; review the code before use.
- Dependency security is not audited; check package.json and build.gradle.kts for dependency versions.
- No rollback or uninstall plan is provided; ensure you understand the impact before installing the plugin.