DOCUMENTATION

RIRD CLI Reference

Computer use agent that automates any browser task with natural language.

Installation Commands Options API Server Batch Mode Examples Memory Server Deploy

1 Installation

RECOMMENDED - One-line install

curl -fsSL https://rird.ai/install.sh | bash

WHAT IT INSTALLS

- RIRD CLI (computer use agent)
- Browser automation engine
- AI brain for task execution

Works on macOS, Linux, and Windows. Requires Node.js 18+.

2 Activation

$ rird

Email: you@company.com

Check your email for a magic link.

Enter code: ABC123

Licensed to you@company.com

A license is required for compute access. View plans.

3 Commands

rird

Interactive mode - type tasks naturally

rird "your task"

Run a single task

rird activate <token>

Activate license key

rird status

Check account, plan, and usage hours

rird update

Check for and install updates

rird examples

Show example prompts

4 Options

Option	Description
--headful	Show browser window (default: headless)
--parallel N	Run N tasks concurrently in batch mode
--server [port]	Start local API server (default: 8765)
--batch <file>	Run tasks from file (one per line)
--schedule <cron> <task>	Schedule recurring task
--fast	Fast extraction mode (no AI reasoning)
--setup	Force re-run setup
--version, -v	Show version
--help, -h	Show help

5 API Server COMING SOON

Start a local HTTP API server for programmatic task submission. Perfect for integrations.

Start the server

rird --server # Runs on port 8765

Endpoints

Method	Path	Description	Auth
POST	/api/task	Submit a task for execution	Bearer token
GET	/api/status	Check license status	Bearer token
GET	/api/health	Health check	None

Example Request

curl -X POST http://localhost:8765/api/task \
  -H "Authorization: Bearer YOUR_LICENSE_KEY" \
  -H "Content-Type: application/json" \
  -d '[object Object]'

Response

{
  "status": "completed",
  "task": "find 10 leads on linkedin for SaaS founders",
  "result": { ... }
}

6 Batch Mode

Run multiple tasks from a file. Use --parallel N for concurrent execution.

Create a tasks file

tasks.txt

# Lines starting with # are ignored
find 10 leads for dental practices
find 10 leads for law firms
find 10 leads for real estate agents

Run sequentially

rird --batch tasks.txt

Run in parallel

rird --batch tasks.txt --parallel 3 # Run 3 tasks concurrently

7 Examples

Lead generation

rird "find 50 SaaS founders on linkedin and save to leads.csv"

Form submission

rird --headful "fill out the contact form on example.com"

Data extraction

rird "extract all product prices from competitor.com/products"

Government data

rird "find DoD contractors on SAM.gov with NAICS 541511"

Scheduled monitoring

rird --schedule "0 9 * * *" "check competitor prices and alert if changed"

8 Login Sessions

Save login sessions to reuse across runs. In interactive mode, type login <service>.

rird> login linkedin

Opening browser for login...

Complete login in the browser window.

Press Enter when done.

Session saved to ~/.rird/browser-profile/

Supported services: gmail, linkedin, twitter, facebook, outlook

9 Memory Management

RIRD maintains a persistent "Internal Github" for your tasks. Every action, result, and context update is version-controlled and stored securely.

Features

✓
Persistent Context: The AI remembers previous tasks, preferences, and findings across sessions.
✓
Version Control: Every task creates a commit in your personal workspace history.
✓
Full Recall: Access detailed JSON logs of any past task via the API.

Accessing Memory

Get current memory context

curl -H "Authorization: Bearer KEY" https://rird.ai/api/user/memory

10 Server Deployment

Run RIRD on your own server or VPS for 24/7 automation. Same install command works anywhere.

1. Install on your server

SSH into your server, then run:

curl -fsSL https://rird.ai/install.sh | bash

2. Activate and run

$ rird activate YOUR_LICENSE_KEY

$ rird --server # Start API server on port 8765

3. Configure residential proxy (recommended)

For production workloads, use a residential proxy service to avoid IP bans. ISP residential proxies with same-city sticky sessions work best for browser automation.

~/.rird/config.yaml

proxy:
  enabled: true
  url: "http://username:password@your-proxy-provider.com:port"
  # ISP residential recommended for stability
  # Same-city proxies reduce latency

Proxy Provider Tips

When choosing a residential proxy service, look for:

- Same-city sticky sessions (consistent location)
- Unlimited or high bandwidth limits
- Predictable monthly pricing
- Fast speeds optimized for browser automation

4. Keep it running

Using PM2 (recommended)

npm install -g pm2
pm2 start "rird --server" --name rird
pm2 save
pm2 startup

VPS Requirements: 2+ CPU cores, 4GB+ RAM, Ubuntu 22.04 or Debian 12. Any cloud provider works (DigitalOcean, Hetzner, Vultr, etc.)

Need Help?

Our team is available to help you get started.

Contact Support View Plans