How to Instantly Enable Claude Code in IntelliJ IDEA with the CC GUI Plugin
This step‑by‑step guide shows how to install, configure, and use the CC GUI plugin to bring Claude Code’s AI coding agent directly into JetBrains IDEA, covering core panels, practical tips, and team‑level best practices for a seamless development workflow.
Many developers who have tried AI‑assisted coding in IDEA find that the powerful models still require frequent terminal switches, breaking the workflow. The CC GUI plugin addresses this by embedding Claude Code and Codex capabilities into the JetBrains sidebar, consolidating conversation, execution, change review, and session management in a single panel.
1. Understanding CC GUI
Claude Code is a command‑line‑centric AI programming agent that can be hard to adopt inside an IDE. CC GUI visualizes its abilities, turning the agent into an in‑IDE workstation.
Launch tasks directly from the IDE.
Perform multi‑file analysis with project context.
Manage Skills, MCP, Git commits, and usage statistics from the panel.
Shift the workflow from "command‑driven" to "interface‑driven".
2. Installation & Configuration
2.1 Install the Plugin
Search for CC GUI in the IntelliJ IDEA plugin marketplace and install the version with the highest download count and most recent updates. After installation, a CC GUI entry appears in the right‑hand toolbar.
2.2 Install Claude Code SDK
The first time you open the panel, you will be prompted to install the Claude Code SDK, which provides the runtime dependencies for the agent. Click the guided install button; the process completes quickly.
2.3 Configure Model Provider
After the SDK is ready, go to Provider Settings . Two typical approaches are supported:
Import an existing configuration by pointing to ~/.claude/settings.json.
Manually add a provider: fill in API key, endpoint URL, and model mapping, then enable as needed.
Once configured, CC GUI can handle everyday development tasks directly inside IDEA.
3. Core Panels Explained
3.1 Skills Panel – Extending Agent Capabilities
Skills are "ability packs". After installing a Skill, you no longer need to specify it manually; describing the task is enough for the system to trigger the appropriate Skill. For example, the web‑access Skill helps with web interaction, login handling, and data scraping.
3.2 MCP Panel – Connecting External Tools
MCP (Model Context Protocol) gives the agent the power to call external tools. The Chrome DevTools MCP integration enables scenarios such as reproducing bugs that require login, cross‑page workflow verification, and UI automation checks.
3.3 Git Panel – Auto‑Generating Commit Messages
The Git panel creates commit messages based on detected changes, helping teams maintain a consistent commit style and reducing repetitive effort.
3.4 Statistics Panel – Monitoring Token Usage
This panel visualizes token consumption, useful for cost management when billing is token‑based. It lets you compare token usage across task types, spot anomalous requests, and forecast budgets to avoid overruns.
4. Practical Tips & Best Practices
4.1 Context‑Aware Queries
With deep IDEA integration, you can ask task‑oriented questions instead of pasting code snippets, such as:
Explain the purpose of the currently selected code.
Generate JavaDoc for the current method.
Analyze performance risks of a Service.
Identify potential NPE locations.
You can also select multiple files and let the agent analyze them in one go.
4.2 Review Diff Before Applying Changes
CC GUI presents suggested modifications in IDEA’s Diff view. Follow this disciplined workflow:
Review changes line by line.
Accept or reject changes in logical blocks.
Apply only after confirming correctness.
This balances AI‑generated speed with human quality control.
4.3 Team Adoption Recommendations
Maintain a shared Skills inventory so new members can align quickly.
Store project‑specific custom Skills in a repository for unified management.
Check for plugin updates monthly to receive new features and bug fixes.
5. Summary
The value of CC GUI lies not in adding another chat window, but in truly embedding Claude Code’s agent capabilities into the main IDEA workflow, reducing context switches and stabilizing AI assistance for project‑level tasks.
If you face network or payment hurdles with the official Claude Code subscription in China, the Code80 service re‑exposes the subscription as a direct API endpoint, making integration smoother for teams or individual developers.
To get started, complete the SDK and provider setup, then run a small end‑to‑end task (ask → modify → review Diff → commit). This quick test will tell you whether the plugin fits your daily development rhythm.
6. About Code80
Code80 aggregates the custom commands, workflow configurations, and deployment scripts the author has refined over a year, offering ready‑to‑copy command templates for Claude Code. Interested readers can follow the project for more tutorials.
7. Final Thoughts
Vibe Coding emphasizes an engineering mindset for managing AI‑assisted programming. Standardized commands such as /commit, /upstream, /progress-save / /progress-load, /deploy, /gitsync, /review, and /bug-add turn repetitive tasks into one‑click operations, letting developers focus on "what to do" while Claude handles "how to do it".
Top Architecture Tech Stack
Sharing Java and Python tech insights, with occasional practical development tool tips.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.
