TypeScript CLI (praisonai-ts)

Complete reference for the PraisonAI TypeScript CLI

PraisonAI TypeScript CLI

The praisonai-ts CLI provides a complete command-line interface for the PraisonAI TypeScript SDK.

Installation

bash

npm install -g praisonai
# or
pnpm add -g praisonai

Quick Start

bash

# Chat with an AI agent
praisonai-ts chat "Hello, how are you?"

# Run with a specific model
praisonai-ts chat "Explain TypeScript" --model openai/gpt-4o

# Get JSON output
praisonai-ts chat "Hello" --json

Global Options

Option	Short	Description
`--verbose`	`-v`	Enable verbose output
`--config`	`-c`	Path to config file
`--profile`	`-p`	Profile name to use
`--output`	`-o`	Output format (json, text, pretty)
`--json`		Shorthand for –output json

Environment Variables

Variable	Description
`OPENAI_API_KEY`	OpenAI API key
`ANTHROPIC_API_KEY`	Anthropic API key
`GOOGLE_API_KEY`	Google API key
`PRAISONAI_MODEL`	Default model

Commands Reference

Core Commands

chat

Chat with an AI agent.

bash

praisonai-ts chat "Your prompt here"
praisonai-ts chat "Write a poem" --stream
praisonai-ts chat "Hello" --model openai/gpt-4o-mini --json

Flag	Description
`--model, -m`	Model to use
`--stream, -s`	Enable streaming
`--session`	Session ID for continuity

run

Run an agent with a task.

bash

praisonai-ts run "Analyze this data"
praisonai-ts run "Build a calculator" --agent my-agent.yaml

Flag	Description
`--agent, -a`	Agent configuration file
`--tools, -t`	Comma-separated tools

workflow

Execute a multi-agent workflow.

bash

praisonai-ts workflow workflow.yaml
praisonai-ts workflow workflow.yaml --parallel

Agent Generation

auto

Auto-generate agents from a topic.

bash

praisonai-ts auto "Build a web scraper"
praisonai-ts auto "Create a data pipeline" --pattern sequential

Flag	Description
`--pattern`	Agent pattern (sequential, parallel, routing)
`--agents`	Number of agents to generate

Specialized Agents

image

Image generation and analysis.

bash

praisonai-ts image generate "A sunset over mountains"
praisonai-ts image analyze https://example.com/image.jpg

Flag	Description
`--size`	Image size (1024×1024, etc.)
`--quality`	Image quality (standard, hd)
`--style`	Image style (vivid, natural)

research

Deep research on a topic.

bash

praisonai-ts research "What are the latest AI trends?"
praisonai-ts research "TypeScript best practices" --depth 5

Flag	Description
`--depth`	Research depth (iterations)
`--max-sources`	Maximum sources to use

query-rewrite

Rewrite queries for better search results.

bash

praisonai-ts query-rewrite "best programming language"
praisonai-ts query-rewrite "how to learn coding" --strategy expand

Flag	Description
`--strategy`	Strategy: expand, simplify, decompose, rephrase, auto

prompt-expand

Expand prompts with more detail.

bash

praisonai-ts prompt-expand "write a story"
praisonai-ts prompt-expand "create an app" --strategy detail

Flag	Description
`--strategy`	Strategy: detail, context, examples, constraints, auto

context

Manage conversation context.

bash

praisonai-ts context chat "What is TypeScript?"
praisonai-ts context summarize "Long text to summarize..."

Flag	Description
`--max-messages`	Maximum messages in context

router

Route requests to appropriate agents.

bash

praisonai-ts router analyze "Help me debug this code"

Safety & Validation

guardrail

Content validation and safety.

bash

praisonai-ts guardrail check "Content to validate"
praisonai-ts guardrail check "Text" --criteria "Must be professional"

Flag	Description
`--criteria`	Custom validation criteria

Memory & State

memory

Manage agent memory.

bash

praisonai-ts memory list
praisonai-ts memory add "Important information"
praisonai-ts memory search "query"
praisonai-ts memory clear

session

Manage agent sessions.

bash

praisonai-ts session list
praisonai-ts session create my-session
praisonai-ts session get my-session
praisonai-ts session delete my-session
praisonai-ts session export my-session

knowledge

Manage knowledge base.

bash

praisonai-ts knowledge list
praisonai-ts knowledge add document.pdf
praisonai-ts knowledge search "query"

Tools & Skills

tools

List or manage tools.

bash

praisonai-ts tools list
praisonai-ts tools info calculator

skills

Manage agent skills.

bash

praisonai-ts skills list
praisonai-ts skills discover ./skills
praisonai-ts skills validate ./my-skill
praisonai-ts skills info my-skill

mcp

Model Context Protocol management.

bash

praisonai-ts mcp list
praisonai-ts mcp add my-server
praisonai-ts mcp start my-server
praisonai-ts mcp stop my-server
praisonai-ts mcp remove my-server

Evaluation

eval

Evaluate agent performance.

bash

praisonai-ts eval accuracy --input "2+2" --expected "4"
praisonai-ts eval performance --iterations 10
praisonai-ts eval reliability --expected-tools "calculator"

Planning

planning

Task planning and todo management.

bash

praisonai-ts planning create "My Project Plan"
praisonai-ts planning todo add "Complete documentation"
praisonai-ts planning todo list

Infrastructure

vector

Vector store management.

bash

praisonai-ts vector info
praisonai-ts vector providers

Available Providers:

MemoryVectorStore
PineconeVectorStore
WeaviateVectorStore
QdrantVectorStore
ChromaVectorStore

db

Database adapter management.

bash

praisonai-ts db info
praisonai-ts db adapters

Available Adapters:

SQLite
Redis (Upstash)
PostgreSQL (Neon)
Memory

cache

Caching management.

bash

praisonai-ts cache info
praisonai-ts cache providers

Available Providers:

MemoryCache
FileCache

graph-rag

Graph-based retrieval augmented generation.

bash

praisonai-ts graph-rag info

reranker

Document reranking.

bash

praisonai-ts reranker info
praisonai-ts reranker providers

Available Providers:

CohereReranker
CrossEncoderReranker
LLMReranker

Monitoring

telemetry

Usage monitoring and analytics.

bash

praisonai-ts telemetry status
praisonai-ts telemetry enable
praisonai-ts telemetry disable
praisonai-ts telemetry clear
praisonai-ts telemetry export

observability

Monitoring and tracing.

bash

praisonai-ts observability info
praisonai-ts observability providers

Available Providers:

ConsoleObservabilityProvider
MemoryObservabilityProvider
LangfuseObservabilityProvider

Voice

voice

Text-to-speech and speech-to-text.

bash

praisonai-ts voice info
praisonai-ts voice providers

Available Providers:

OpenAIVoiceProvider
ElevenLabsVoiceProvider

Agent Handoff

handoff

Agent handoff management.

bash

praisonai-ts handoff info

Utility Commands

providers

List available LLM providers.

bash

praisonai-ts providers
praisonai-ts providers --json

version

Show CLI version.

bash

praisonai-ts version
praisonai-ts version --json

help

Show help information.

bash

praisonai-ts help
praisonai-ts help chat
praisonai-ts help --json

JSON Output Format

All commands support --json flag for structured output:

Success Response

json

{
  "success": true,
  "data": { ... },
  "meta": {
    "duration_ms": 1234,
    "model": "openai/gpt-4o-mini",
    "tokens": { "input": 10, "output": 50 }
  }
}

Error Response

json

{
  "success": false,
  "error": {
    "code": "MISSING_ARGUMENT",
    "message": "Please provide a prompt"
  }
}

Exit Codes

Code	Description
0	Success
1	Runtime error
2	Invalid arguments
3	Config error
4	Network error
5	Auth error

Complete Command List

Command	Description
`chat`	Chat with an AI agent
`run`	Run an agent with a task
`workflow`	Execute a multi-agent workflow
`auto`	Auto-generate agents from topic
`image`	Image generation and analysis
`research`	Deep research on a topic
`query-rewrite`	Rewrite queries for better search
`prompt-expand`	Expand prompts with more detail
`context`	Manage conversation context
`router`	Route requests to agents
`guardrail`	Content validation and safety
`memory`	Manage agent memory
`session`	Manage agent sessions
`knowledge`	Manage knowledge base
`tools`	List or manage tools
`skills`	Manage agent skills
`mcp`	Model Context Protocol management
`eval`	Evaluate agent performance
`planning`	Task planning and todo management
`vector`	Vector store management
`db`	Database adapter management
`cache`	Caching management
`graph-rag`	Graph-based RAG
`reranker`	Document reranking
`telemetry`	Usage monitoring
`observability`	Monitoring and tracing
`voice`	Text-to-speech and speech-to-text
`handoff`	Agent handoff management
`providers`	List LLM providers
`version`	Show CLI version
`help`	Show help information