Skip to main content

​
Prerequisites

  • Java 21+ on PATH
  • Claude Code installed and authenticated (claude.ai/code)
  • JetBrains IDE 2026.1+ with AI Assistant plugin enabled

​
Install from Release

curl -L https://github.com/markpollack/bud/releases/download/v0.1.0/bud-acp-server-0.1.0-unix.tar.gz | tar xz
cd bud-acp-server-0.1.0

​
Install from Source

git clone https://github.com/markpollack/bud.git
cd bud
./mvnw package -DskipTests
This produces:
  • bud-acp-server/target/bud-acp-server-0.1.0-jar-with-dependencies.jar — the ACP agent
  • bud-mcp-server/target/bud-mcp-server-0.1.0.jar — the MCP tool server

​
Configure JetBrains

Create or edit ~/.jetbrains/acp.json: 
{
  "default_mcp_settings": {},
  "agent_servers": {
    "Bud": {
      "command": "java",
      "args": [
        "-Dbud.mcp.jar=/path/to/bud/bud-mcp-server/target/bud-mcp-server-0.1.0.jar",
        "-jar",
        "/path/to/bud/bud-acp-server/target/bud-acp-server-0.1.0-jar-with-dependencies.jar"
      ]
    }
  }
}
Replace /path/to/bud with the absolute path to your cloned repository. The -Dbud.mcp.jar system property tells the ACP agent where to find the MCP tool server.
Restart your JetBrains IDE after editing the config.

​
Your First Project

  1. Open any project in IntelliJ
  2. Open AI Chat tool window
  3. Select “Bud” from the agent dropdown
  4. Type: create a REST app called hello-api in /tmp
Bud will scaffold a complete Spring Boot REST project with:
  • GreetingController with GET endpoint
  • GreetingService with business logic
  • ProblemDetail error handling
  • Working tests
  • Maven wrapper

​
Verify

cd /tmp/hello-api
./mvnw test

​
Try More Commands

add actuator to /tmp/hello-api

analyze the project at /tmp/hello-api

list available project templates

​
How It Works

When you send a message to Bud in AI Chat:
  1. JetBrains sends the prompt to Bud’s ACP agent over stdio
  2. Bud forwards the goal to Claude Code CLI with its 16 MCP tools registered
  3. Claude Code picks the right tools (e.g., create_project, add_actuator)
  4. The tools execute deterministically — no LLM involved in the actual scaffolding
  5. Results stream back through ACP to your chat window
The first prompt may take 30-60 seconds (CLI startup + MCP server startup + LLM reasoning). Subsequent prompts in the same session are faster.