Developer Documentation

Build with Hilla.

Web app, MCP server, CLI, and API documentation. Everything you need to plan on the canvas and integrate Hilla into your workflow.

Getting Started

Overview

Hilla is an AI co-founder that turns one sentence into a full visual project board. Plan, build, track, and ship on the same surface. You can interact with Hilla through the web app, CLI, or any AI tool that supports the Model Context Protocol.

The primary surface is the web canvas at hilla.ai. This documentation also covers the MCP server for AI-native integrations, the REST API for programmatic access, and the CLI for power-user terminal workflows.

Getting Started

Quick Start

Go to hilla.ai and describe what you want to build. Hilla asks a few clarifying questions, then generates a full spatial board with cards, tasks, dependencies, and sprint plans in about 30 seconds.

From there, work directly on the canvas: drag cards, edit tasks, embed live widgets from your stack (Vercel, Stripe, PostHog, Sentry), and share the board with collaborators. The canvas is the primary surface for planning and tracking.

For AI-native workflows, connect an MCP-compatible assistant (see below). For terminal scripting, install the CLI.

MCP Server

What is MCP?

The Model Context Protocol is an open standard that lets AI tools connect to external services. Hilla runs an MCP server so your AI assistant can create, read, and manage cards on your board directly, without switching context.

When you connect Hilla via MCP, your AI tool gains access to your project boards, tasks, and sprint plans. It can suggest tasks, update progress, and even execute work on your behalf.

MCP Server

Setup

Hilla runs a remote MCP server over HTTP/SSE. No local process needed. Generate an API token in Settings → API Tokens (tokens start with play_), then add the server to your AI tool's config file.

For Claude Desktop, add to claude_desktop_config.json:

claude_desktop_config.json
{
  "mcpServers": {
    "hilla": {
      "type": "sse",
      "url": "https://api.hilla.ai/functions/v1/mcp-server",
      "headers": {
        "Authorization": "Bearer play_YOUR_TOKEN_HERE"
      }
    }
  }
}

For Cursor, Windsurf, or Cline, add to .cursor/mcp.json (or the equivalent for your editor):

.cursor/mcp.json
{
  "mcpServers": {
    "hilla": {
      "type": "sse",
      "url": "https://api.hilla.ai/functions/v1/mcp-server",
      "headers": {
        "Authorization": "Bearer play_YOUR_TOKEN_HERE"
      }
    }
  }
}

For full setup guides including ChatGPT and claude.ai, see /docs/mcp.

MCP Server

Available Tools

The MCP server exposes these tools to connected AI clients. Each tool maps to a Hilla board operation.

Tool
Description
list_projects
List all your projects
get_board
Get full board state for a project
get_card
Get card details with tasks
get_progress
Get progress summary
create_project
Create a new project
create_card
Add a card to the board
add_tasks
Add tasks to a card
complete_task
Mark a task as done
update_card
Update card title, description, category, priority, or position
set_card_priority
Set card priority
delete_card
Remove a card from the board
search
Search projects, cards, and tasks

For the full reference including credit costs, see /docs/mcp.

MCP Server

Supported Clients

Any AI tool that implements the Model Context Protocol can connect to Hilla. These clients have been tested and confirmed working:

C

Claude Code

Anthropic’s CLI agent. Add Hilla to your project’s MCP config.

Cu

Cursor

AI code editor. Configure in Settings → MCP Servers.

G

ChatGPT

OpenAI’s assistant. Connect via the MCP plugin system.

W

Windsurf

Codeium’s editor. Add to your MCP configuration file.

CLI

Command Line Interface

The CLI is a power-user companion for developers who want to script, pipe, and automate board operations from the terminal. The web canvas is the primary surface; the CLI extends it.

Every command supports --json output for scripting and automation. Use hilla help <command> for detailed usage.

CLI

Installation

Install the Hilla CLI globally via npm. Requires Node.js 18 or later.

terminal
$ npm install -g @hilla/cli

$ hilla --version
hilla/1.0.0

Authenticate with your Hilla account:

terminal
$ hilla auth login
→ Opening browser for authentication...
✓ Authenticated as you@example.com

CLI

Commands

Command
Description
hilla plan <prompt>
Generate a full project plan from a description
hilla board [id]
Open or list project boards
hilla tasks
List all tasks in the current project
hilla task <id>
View details for a specific task
hilla exec <id>
Execute a task with AI assistance
hilla export [format]
Export board as Markdown, JSON, or to Linear/GitHub
hilla auth login
Authenticate with your Hilla account
hilla auth logout
Sign out of the current session
hilla config
View or update CLI configuration

Chain commands with standard Unix pipes:

terminal
$ hilla export --json | jq '.tasks[] | select(.sprint == 1)'

CLI

Configuration

The CLI stores configuration in ~/.hilla/config.json. You can edit it directly or use hilla config to update values.

Key
Default
Description
defaultFormat
markdown
Export format (markdown, json)
editor
$EDITOR
Preferred editor for task editing
apiUrl
https://api.hilla.ai
API endpoint
color
true
Enable colored output

API Reference

Authentication

All API requests require a bearer token. Generate one from your account settings or via the CLI.

terminal
$ hilla auth token
→ Your API token: hilla_sk_...

# Use in requests:
$ curl -H "Authorization: Bearer hilla_sk_..." \
    https://api.hilla.ai/v1/projects

Tokens don't expire but can be revoked from your account settings at any time.

API Reference

Projects

Create and manage project boards programmatically.

Method
Endpoint
Description
GET
/v1/projects
List all projects
POST
/v1/projects
Create a new project
GET
/v1/projects/:id
Get project details
PATCH
/v1/projects/:id
Update project settings
DELETE
/v1/projects/:id
Delete a project
terminal
$ curl -X POST https://api.hilla.ai/v1/projects \
    -H "Authorization: Bearer hilla_sk_..." \
    -H "Content-Type: application/json" \
    -d '{"prompt": "Build a SaaS for restaurant analytics"}'

{
  "id": "proj_rest_analytics",
  "name": "Restaurant Analytics SaaS",
  "tasks": 47,
  "sprints": 3,
  "url": "https://hilla.ai/b/rest-analytics"
}

API Reference

Tasks

Read and manage individual task cards on a project board.

Method
Endpoint
Description
GET
/v1/projects/:id/tasks
List tasks (filterable by sprint, status)
POST
/v1/projects/:id/tasks
Create a task
GET
/v1/tasks/:id
Get task details
PATCH
/v1/tasks/:id
Update a task
POST
/v1/tasks/:id/execute
Execute a task with AI
DELETE
/v1/tasks/:id
Delete a task
terminal
$ curl https://api.hilla.ai/v1/projects/proj_rest_analytics/tasks?sprint=1 \
    -H "Authorization: Bearer hilla_sk_..."

{
  "tasks": [
    {
      "id": "task_001",
      "title": "Auth system (OAuth + magic link)",
      "sprint": 1,
      "status": "todo",
      "dependencies": []
    },
    ...
  ]
}
Get started

Ready to build?

Install the CLI and generate your first project board in under 30 seconds. Free to start.

$ npm install -g @hilla/cli