Skip to main content

​
Quick Start

Get started with AnyTask in 4 simple steps.

​
Step 1: Create Your Project

Connect GitHub and set up your workspace.

​
Connect GitHub

  1. Go to anyt.dev and sign in
  2. Click Connect GitHub to link your GitHub account
  3. Authorize AnyTask to access your repositories

​
Create Project

  1. Select the repository you want to work with
  2. Choose a workspace name and identifier
  3. Click Create Project
Once completed, you’ll see: “Project created! You’re ready to start working.”

​
Step 2: Setup CLI

Install the AnyTask CLI and authenticate.

​
Install CLI

curl -fsSL https://anyt.dev/install.sh | sh

# Using uvx (recommended - no installation needed)
uvx anyt --help

# Using pipx (install once, use everywhere)
pipx install anyt

# Using pip
pip install anyt

​
Get API Key

  1. Visit anyt.dev/home/settings/api-keys
  2. Click Get API Key
  3. Copy the generated key (starts with anyt_agent_)

​
Login

Run the login command and paste your API key when prompted: 
anyt login
This will:
  • Validate your API key
  • Store it securely for future use
  • Configure the CLI to connect to your workspace
Verify your setup: 
anyt health check

​
Step 3: Work with Your Favorite Coding Agent

Use slash commands to plan and execute tasks with your preferred AI coding agent.

​
Install Agent Commands

Install slash commands for your coding agents: 
anyt agent install
This will interactively let you select which agents to install commands for. You can also specify agents directly: 
# Install for specific agents
anyt agent install claude
anyt agent install cursor
anyt agent install gemini
anyt agent install codex

​
Supported Coding Agents

AgentCommand FormatDescription
ClaudeMarkdownClaude Code CLI integration
CursorMarkdownCursor IDE integration
GeminiTOMLGemini CLI integration
CodexYAMLOpenAI Codex CLI integration

​
Available Slash Commands

Once installed, you can use these slash commands in your coding agent:
CommandDescription
/anyt:anyt-new {requirement}Create a new task in AnyTask cloud
/anyt:anyt-nextPick the next suggested task to work on
/anyt:anyt-work {TASK_ID}Continue working on a specific task

​
Example Workflow

  1. Create a task using the slash command in your coding agent: 
    /anyt:anyt-new Add user authentication with OAuth support
  2. Pick the next task to work on: 
    /anyt:anyt-next
  3. Continue working on a specific task: 
    /anyt:anyt-work DEV-42

​
Step 4: Setup Background Coding Agent

Enable your favorite AI coding agent to work autonomously on tasks.

​
1. Enable Your Coding Agent

Set up which coding agents should work on your tasks:
  1. Go to your project settings in anyt.dev
  2. Click Setup Coding Agents
  3. Enable the agents you want to use (Claude, Cursor, Gemini, Codex)

​
2. Assign a Task to Your Agent

Create a task and select your agent as the assignee: 
anyt task add "Implement feature X" --status todo
Assign it to your coding agent via the web UI, or let the worker pick it up automatically.

​
3. Start the Background Worker

Run the worker to let your AI agent process tasks autonomously: 
anyt worker start

# Interactive workflow selection
anyt worker start

# Specify workflow directly
anyt worker start --workflow local_dev
anyt worker start --workflow remote_dev

# With custom poll interval
anyt worker start --poll-interval 10

# Scope to specific project
anyt worker start --project-id 123
What the worker does:
  • Polls for available tasks assigned to your agent
  • Matches tasks to appropriate workflows
  • Executes implementation using your coding agent
  • Updates task status and adds comments
  • Creates pull requests (for remote_dev workflow)

​
Common Commands

​
Task Management

# Create tasks
anyt task add "Task title" --priority 1 --status todo

# List tasks
anyt task list
anyt task list --status todo

# View task details
anyt task show DEV-123

# Update tasks
anyt task edit DEV-123 --status in_progress

# Mark as done
anyt task done DEV-123

# Delete tasks
anyt task rm DEV-123 --force

​
Active Task Workflow

# Pick a task to work on
anyt task pick DEV-123

# View active task
anyt active

# Mark active task as done
anyt task done

# Add comment
anyt comment add -m "Completed implementation"

​
Dependencies

# Add dependency
anyt task dep add DEV-43 --on DEV-42

# List dependencies
anyt task dep list DEV-43

# Remove dependency
anyt task dep rm DEV-43 --on DEV-42

​
Quick Reference

CommandDescription
anyt loginAuthenticate with AnyTask
anyt initInitialize workspace in current directory
anyt task listList all tasks
anyt task add "title"Create a new task
anyt task pickPick a task to work on
anyt task doneMark active task as done
anyt agent installInstall coding agent commands
anyt worker startStart background worker
anyt self-updateUpdate CLI to latest version

​
Next Steps

CLI Reference

Complete command documentation

Worker System

Automated task execution

​
Troubleshooting

​
Command not found

# Check if CLI is installed
which anyt

# Add to PATH if needed
export PATH="$HOME/.local/bin:$PATH"

​
Authentication issues

# Re-login
anyt login

# Check API connectivity
anyt health check

​
Not in a workspace directory

# Initialize workspace
anyt init

​
Getting Help

# General help
anyt --help

# Command-specific help
anyt task --help
anyt task add --help
For more help: [email protected]