Skip to content

CLI Reference

[![Version](https://img.shields.io/badge/version-2.0.0-blue.svg)](https://github.com/isathish/agenticaiframework) [![Modules](https://img.shields.io/badge/modules-400%2B-brightgreen.svg)](https://github.com/isathish/agenticaiframework) [![Enterprise](https://img.shields.io/badge/enterprise-237%20modules-gold.svg)](https://github.com/isathish/agenticaiframework) [![CLI](https://img.shields.io/badge/CLI-Command--Line-success.svg)](https://github.com/isathish/agenticaiframework)

Enterprise CLI

Part of 237 enterprise modules with comprehensive CLI tools. See Enterprise Documentation.

Overview

Complete command-line interface (CLI) reference for AgenticAI Framework. This guide covers all CLI commands, options, and usage examples for managing agents, tasks, and configurations.

Installation

Bash
1
2
3
4
5
6
7
8
# Install AgenticAI with CLI tools
pip install agenticaiframework[cli]

# Verify installation
agenticai --version

# Show help
agenticai --help

Global Options

Options available for all commands:

Option Short Description Default
--version -v Show version -
--help -h Show help message -
--config -c Path to config file agenticai.yaml
--verbose -V Enable verbose output false
--quiet -q Suppress output false
--log-level -l Set log level INFO
--output -o Output format (json, yaml, table) table

Usage

Bash
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
# Show version
agenticai --version

# Use custom config
agenticai --config /path/to/config.yaml agent list

# Enable verbose mode
agenticai --verbose agent create --name my_agent

# JSON output
agenticai --output json agent list

Agent Commands

agenticai agent create

Create a new agent.

Bash
1
2
3
4
5
6
agenticai agent create \
  --name "research_agent" \
  --role "researcher" \
  --capabilities "search,analyze,summarize" \
  --model "gpt-4" \
  --max-tokens 4000

Options:

Option Required Type Description
--name Yes string Agent name
--role Yes string Agent role
--capabilities No list Comma-separated capabilities
--model No string LLM model to use
--max-tokens No int Maximum tokens per request
--temperature No float Sampling temperature (0-2)
--config No string Path to agent config file

Example:

Bash
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
# Create simple agent
agenticai agent create --name analyst --role "data_analyst"

# Create agent with config file
agenticai agent create --config agents/research_agent.yaml

# Create agent with capabilities
agenticai agent create \
  --name coder \
  --role "developer" \
  --capabilities "code_generation,code_review,testing"

agenticai agent list

List all agents.

Bash
1
agenticai agent list [OPTIONS]

Options:

Option Type Description
--status string Filter by status (active, paused, stopped)
--role string Filter by role
--format string Output format (table, json, yaml)
--limit int Limit number of results

Example:

Bash
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
# List all agents
agenticai agent list

# List active agents
agenticai agent list --status active

# List in JSON format
agenticai agent list --format json

# Limit results
agenticai agent list --limit 10

agenticai agent show

Show detailed information about an agent.

Bash
1
agenticai agent show <agent_id> [OPTIONS]

Options:

Option Type Description
--format string Output format
--show-history bool Include execution history
--show-stats bool Include statistics

Example:

Bash
1
2
3
4
5
6
7
8
# Show agent details
agenticai agent show agent_001

# Show with history
agenticai agent show agent_001 --show-history

# JSON output
agenticai agent show agent_001 --format json

agenticai agent start

Start an agent.

Bash
1
agenticai agent start <agent_id>

Example:

Bash
1
agenticai agent start agent_001

agenticai agent stop

Stop an agent.

Bash
1
agenticai agent stop <agent_id> [--force]

Options:

Option Type Description
--force bool Force stop without graceful shutdown

Example:

Bash
1
2
3
4
5
# Graceful stop
agenticai agent stop agent_001

# Force stop
agenticai agent stop agent_001 --force

agenticai agent delete

Delete an agent.

Bash
1
agenticai agent delete <agent_id> [--confirm]

Options:

Option Type Description
--confirm bool Skip confirmation prompt

Example:

Bash
1
2
3
4
5
# Delete with confirmation
agenticai agent delete agent_001

# Delete without confirmation
agenticai agent delete agent_001 --confirm

agenticai task create

Create and execute a task.

Bash
1
2
3
4
5
6
agenticai task create \
  --name "Data Analysis" \
  --description "Analyze sales data" \
  --agent agent_001 \
  --priority 5 \
  --timeout 60

Options:

Option Required Type Description
--name Yes string Task name
--description No string Task description
--agent Yes string Agent ID to execute task
--priority No int Priority (1-10)
--timeout No int Timeout in seconds
--input No string Input data (JSON string)
--input-file No string Path to input file

Example:

Bash
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
# Create basic task
agenticai task create \
  --name "Analysis" \
  --agent agent_001

# Create with input data
agenticai task create \
  --name "Process Data" \
  --agent agent_001 \
  --input '{"data": [1,2,3,4,5]}'

# Create from file
agenticai task create \
  --name "Batch Processing" \
  --agent agent_001 \
  --input-file data/input.json

agenticai task list

List all tasks.

Bash
1
agenticai task list [OPTIONS]

Options:

Option Type Description
--status string Filter by status (pending, running, completed, failed)
--agent string Filter by agent ID
--priority int Filter by priority
--limit int Limit results

Example:

Bash
1
2
3
4
5
6
7
8
# List all tasks
agenticai task list

# List running tasks
agenticai task list --status running

# List tasks for specific agent
agenticai task list --agent agent_001

agenticai task show

Show task details.

Bash
1
agenticai task show <task_id> [OPTIONS]

Options:

Option Type Description
--format string Output format
--show-output bool Include task output
--show-logs bool Include execution logs

Example:

Bash
1
2
3
4
5
6
7
8
# Show task details
agenticai task show task_001

# Show with output
agenticai task show task_001 --show-output

# Show with logs
agenticai task show task_001 --show-logs

agenticai task cancel

Cancel a running task.

Bash
1
agenticai task cancel <task_id>

Example:

Bash
1
agenticai task cancel task_001

agenticai task retry

Retry a failed task.

Bash
1
agenticai task retry <task_id>

Example:

Bash
1
agenticai task retry task_001

agenticai memory store

Store data in memory.

Bash
1
2
3
4
agenticai memory store \
  --key "user_profile_123" \
  --value '{"name": "John", "email": "john@example.com"}' \
  --ttl 3600

Options:

Option Required Type Description
--key Yes string Memory key
--value Yes string Value (JSON string)
--value-file No string Path to value file
--ttl No int Time-to-live (seconds)
--tags No list Comma-separated tags

Example:

Bash
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
# Store simple value
agenticai memory store --key config --value '{"setting": "value"}'

# Store from file
agenticai memory store --key data --value-file data.json

# Store with TTL and tags
agenticai memory store \
  --key session_123 \
  --value '{"user": "john"}' \
  --ttl 1800 \
  --tags "session,user"

agenticai memory get

Retrieve data from memory.

Bash
1
agenticai memory get <key> [OPTIONS]

Options:

Option Type Description
--format string Output format
--output-file string Save to file

Example:

Bash
1
2
3
4
5
# Get value
agenticai memory get user_profile_123

# Save to file
agenticai memory get user_profile_123 --output-file profile.json

agenticai memory delete

Delete data from memory.

Bash
1
agenticai memory delete <key>

Example:

Bash
1
agenticai memory delete user_profile_123

agenticai memory list

List all memory entries.

Bash
1
agenticai memory list [OPTIONS]

Options:

Option Type Description
--tags list Filter by tags
--pattern string Key pattern to match
--limit int Limit results

Example:

Bash
1
2
3
4
5
6
7
8
# List all
agenticai memory list

# Filter by tags
agenticai memory list --tags "session,active"

# Match pattern
agenticai memory list --pattern "user_*"

agenticai memory clear

Clear all memory entries.

Bash
1
agenticai memory clear [--confirm]

Example:

Bash
1
2
3
4
5
# Clear with confirmation
agenticai memory clear

# Clear without confirmation
agenticai memory clear --confirm

agenticai llm generate

Generate text using LLM.

Bash
1
2
3
4
5
agenticai llm generate \
  --prompt "Explain quantum computing" \
  --model "gpt-4" \
  --max-tokens 500 \
  --temperature 0.7

Options:

Option Type Description
--prompt string Prompt text
--prompt-file string Path to prompt file
--model string Model to use
--max-tokens int Maximum tokens
--temperature float Temperature (0-2)
--stream bool Stream response

Example:

Bash
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
# Simple generation
agenticai llm generate --prompt "Write a haiku about AI"

# From file
agenticai llm generate --prompt-file prompts/analysis.txt

# With streaming
agenticai llm generate \
  --prompt "Tell a story" \
  --stream

agenticai llm models

List available models.

Bash
1
agenticai llm models [--provider openai]

Example:

Bash
1
2
3
4
5
# List all models
agenticai llm models

# List OpenAI models
agenticai llm models --provider openai

Show system status.

Bash
1
agenticai monitor status

Example output:

Text Only
1
2
3
4
5
6
7
System Status
─────────────────────────────
Status: Healthy
Agents: 15 active, 3 paused
Tasks: 45 running, 120 completed
Memory Usage: 2.4 GB / 8 GB
CPU Usage: 45%

agenticai monitor metrics

Show detailed metrics.

Bash
1
agenticai monitor metrics [OPTIONS]

Options:

Option Type Description
--interval int Refresh interval (seconds)
--follow bool Continuous monitoring
--metrics list Specific metrics to show

Example:

Bash
1
2
3
4
5
6
7
8
# Show metrics once
agenticai monitor metrics

# Continuous monitoring
agenticai monitor metrics --follow --interval 5

# Specific metrics
agenticai monitor metrics --metrics "cpu,memory,tasks"

agenticai monitor logs

View application logs.

Bash
1
agenticai monitor logs [OPTIONS]

Options:

Option Type Description
--follow bool Follow log output
--lines int Number of lines to show
--level string Log level filter
--agent string Filter by agent ID

Example:

Bash
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
# View recent logs
agenticai monitor logs --lines 100

# Follow logs
agenticai monitor logs --follow

# Filter by level
agenticai monitor logs --level ERROR --follow

# Agent-specific logs
agenticai monitor logs --agent agent_001 --follow

Config Commands

agenticai config show

Show current configuration.

Bash
1
agenticai config show [--section agents]

Example:

Bash
1
2
3
4
5
6
7
8
# Show all config
agenticai config show

# Show specific section
agenticai config show --section agents

# JSON output
agenticai config show --format json

agenticai config set

Set configuration value.

Bash
1
agenticai config set <key> <value>

Example:

Bash
1
2
3
4
5
# Set value
agenticai config set agents.max_agents 100

# Set nested value
agenticai config set llm.default_model gpt-4-turbo

agenticai config get

Get configuration value.

Bash
1
agenticai config get <key>

Example:

Bash
1
2
3
4
5
# Get value
agenticai config get agents.max_agents

# Get nested value
agenticai config get llm.default_model

agenticai config validate

Validate configuration.

Bash
1
agenticai config validate [--config config.yaml]

Example:

Bash
1
2
3
4
5
# Validate current config
agenticai config validate

# Validate specific file
agenticai config validate --config staging-config.yaml

Run test suite.

Bash
1
agenticai test run [OPTIONS]

Options:

Option Type Description
--module string Specific module to test
--coverage bool Generate coverage report
--verbose bool Verbose output

Example:

Bash
1
2
3
4
5
6
7
8
# Run all tests
agenticai test run

# Run specific module
agenticai test run --module agents

# With coverage
agenticai test run --coverage

agenticai test benchmark

Run performance benchmarks.

Bash
1
agenticai test benchmark [--suite performance]

Example:

Bash
1
2
3
4
5
# Run all benchmarks
agenticai test benchmark

# Run specific suite
agenticai test benchmark --suite performance

agenticai init

Initialize a new project.

Bash
1
agenticai init [project_name] [OPTIONS]

Options:

Option Type Description
--template string Project template (basic, advanced, custom)
--with-examples bool Include example code

Example:

Bash
1
2
3
4
5
6
7
8
# Initialize project
agenticai init my_project

# With template
agenticai init my_project --template advanced

# With examples
agenticai init my_project --with-examples

agenticai doctor

Diagnose system issues.

Bash
1
agenticai doctor

Example output:

Text Only
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
Running diagnostics...

 Python version: 3.11.5
 Dependencies: All installed
 Configuration: Valid
 API keys: Configured
 Redis connection: Failed (Connection refused)
 Disk space: Low (15% remaining)

2 issues found, 4 checks passed

agenticai version

Show version information.

Bash
1
agenticai version [--full]

Example:

Bash
1
2
3
4
5
# Short version
agenticai version

# Full version info
agenticai version --full
Bash
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
# 1. Initialize project
agenticai init my_agent_app --with-examples

# 2. Create agent
agenticai agent create \
  --name research_agent \
  --role researcher \
  --model gpt-4

# 3. Create and run task
agenticai task create \
  --name "Research AI trends" \
  --agent research_agent \
  --input '{"topic": "artificial intelligence", "year": 2024}'

# 4. Monitor execution
agenticai monitor logs --follow --agent research_agent

# 5. Check results
agenticai task show task_001 --show-output

# 6. View metrics
agenticai monitor metrics

# 7. Stop agent
agenticai agent stop research_agent

Batch Operations

Bash
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
# Create multiple agents
for i in {1..5}; do
  agenticai agent create --name "agent_$i" --role worker
done

# List all agents
agenticai agent list --format json | jq '.[] | .id'

# Stop all agents
agenticai agent list --format json | jq -r '.[] | .id' | xargs -I {} agenticai agent stop {}

Shell Completion

Bash
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
# Bash
agenticai --install-completion bash
source ~/.bashrc

# Zsh
agenticai --install-completion zsh
source ~/.zshrc

# Fish
agenticai --install-completion fish

Back to Top