Skip to main content
This guide walks you through setting up Docbot locally from installation to running your first documentation task.

​
Prerequisites

Before installing Docbot, ensure you have the following tools installed:
  • Bun - Required runtime (will not change)
  • Docker - For running Qdrant locally
  • ripgrep (rg) - For fast exact-match search
  • AI Gateway API key - Vercel AI Gateway access
You can check if ripgrep is installed by running rg --version in your terminal.

​
Installation

1

Install Docbot

You can run Docbot directly with bunx without installing globally:
bunx @helmlabs/docbot --help
Or install globally for easier access:
bun add -g @helmlabs/docbot
docbot --help
2

Set up environment variables

Create a .env file in your project root and add your AI Gateway API key:
AI_GATEWAY_API_KEY=your_api_key_here
Never commit your .env file to version control. Add it to .gitignore immediately.
3

Initialize Docbot in your project

Navigate to your project directory and run the initialization command:
bunx @helmlabs/docbot init
This command will:
  • Create a .docbot/ directory for cache and local state
  • Generate docbot.config.jsonc with default configuration
  • Add .docbot/ to your .gitignore
  • Check for Docker and attempt to start Qdrant automatically
Use --force to overwrite existing configuration, or --skip-docker if you’re setting up Qdrant manually.
4

Configure your project paths

Edit docbot.config.jsonc to specify your documentation and codebase paths:
{
  "projectSlug": "my-project",
  "paths": {
    "docs": "./docs",
    "codebase": ["./src", "./packages/shared"]
  },
  "qdrant": {
    "url": "http://127.0.0.1:6333",
    "collections": {
      "docs": "docbot_my-project_docs",
      "code": "docbot_my-project_code"
    }
  }
}
The codebase field accepts either a single string or an array of paths for monorepo support.
5

Start Qdrant (if not auto-started)

If Qdrant wasn’t automatically started during initialization, start it manually:
docker run --rm -p 6333:6333 -v "$(pwd)/.docbot/qdrant_storage:/qdrant/storage" qdrant/qdrant
Verify Qdrant is running by checking the health endpoint:
curl http://127.0.0.1:6333/health
You should receive a response indicating the service is healthy.
6

Index your documentation and codebase

Build the search index from your docs and code:
bunx @helmlabs/docbot index
This command will:
  • Read all documentation files from your docs path
  • Parse and chunk code from your codebase paths
  • Generate embeddings using the configured model
  • Store vectors in Qdrant collections
Indexing can take 5-10 minutes depending on the size of your codebase and documentation. This is a one-time operation unless you use --force to rebuild.
7

Run your first task

Once indexing is complete, run Docbot with a documentation task:
bunx @helmlabs/docbot run "document the authentication flow"
Docbot will:
  1. Analyze your docs and codebase
  2. Propose a structured plan
  3. Wait for your approval
  4. Execute the changes
  5. Present a review

​
Verify your setup

Test that everything is working correctly: 
# Test search functionality
bunx @helmlabs/docbot search "authentication" --type hybrid --limit 5

# Check Qdrant collections
curl http://127.0.0.1:6333/collections

​
Next steps

​
Common setup options

​
Running without the TUI

If you prefer to run the server without the terminal UI: 
bunx @helmlabs/docbot serve --port 3070

​
Using a remote Qdrant instance

Update your docbot.config.jsonc to point to a remote instance: 
{
  "qdrant": {
    "url": "https://your-qdrant-instance.com"
  }
}

​
Force re-indexing

To rebuild your index from scratch: 
bunx @helmlabs/docbot index --force

Build docs developers (and LLMs) love

Get started for freeTalk to us