import { Callout } from 'fumadocs-ui/components/callout'; import { Tabs, Tab } from 'fumadocs-ui/components/tabs';

Installation

Install the CLI globally

go install github.com/nuulab/goflow/cmd/cli@latest

Service Commands

Start GoFlow services for development or production.

Start the API server

# Start on default port 8080
goflow server
 
# Custom port
goflow server -p 3000
 
# With verbose logging
goflow server -p 8080 -v

Custom queue name

goflow worker -q high-priority -c 5

  </Tab>
  <Tab value="Scheduler">
```bash title="Start the scheduler"
# Start the cron scheduler
goflow scheduler

Agent Commands

Run AI agents from the command line.

Run an agent task

# Single task
goflow agent "What is the weather in Tokyo?"
 
# Interactive chat mode
goflow agent -i
 
# With specific tools enabled
goflow agent -t calculator,web_search "Calculate 25% of 400"
 
# Custom model
goflow agent --provider anthropic --model claude-3-opus "Write a poem"

Agent Flags

Flag	Description	Default
`-i, --interactive`	Interactive chat mode	`false`
`-t, --tools`	Comma-separated tool list	all
`--provider`	LLM provider	`openai`
`--model`	Model name	provider default
`-m, --max-iterations`	Max reasoning steps	`10`

Job Commands

Manage background jobs in the queue.

Job operations

# List all jobs
goflow job list
 
# Filter by status
goflow job list -s pending
goflow job list -s failed
 
# Enqueue a new job
goflow job enqueue -t send_email -d '{"to":"user@example.com"}'
 
# Get job details
goflow job status job-abc123
 
# Retry a failed job
goflow job retry job-abc123

Workflow Commands

Control workflow executions.

Workflow operations

# List all workflows
goflow workflow list
 
# Start a workflow with input
goflow workflow start order-process -i '{"order_id":123}'
 
# Check workflow status
goflow workflow status wf-abc123
 
# Pause a running workflow
goflow workflow pause wf-abc123
 
# Resume a paused workflow
goflow workflow resume wf-abc123
 
# Send a signal
goflow workflow signal wf-abc123 approval '{"approved":true}'

Queue Commands

Monitor queue statistics and events.

Queue monitoring

# Show queue statistics
goflow queue stats
 
# Watch events in real-time
goflow events -f
 
# Filter events by job
goflow events -j job-abc123

Dead Letter Queue

Manage failed jobs in the DLQ.

DLQ operations

# List DLQ entries
goflow dlq list
 
# Retry a specific job
goflow dlq retry job-abc123
 
# Retry all DLQ jobs
goflow dlq retry-all
 
# Purge the DLQ
goflow dlq purge

Global Flags

Flag	Description	Default
`--config`	Config file path	`./goflow.yaml`
`--redis`	Redis address	`localhost:6379`
`-v, --verbose`	Verbose output	`false`

Configuration File

Create a goflow.yaml file for persistent configuration:

goflow.yaml

# API Server
server:
  port: 8080
  host: "0.0.0.0"
 
# Redis/DragonflyDB
redis: localhost:6379
 
# Agent defaults
agent:
  max_iterations: 10
  default_provider: openai
 
# LLM Configuration
llm:
  provider: openai
  model: gpt-4o
  api_key: ${OPENAI_API_KEY}  # Use env var

CLI Reference