PSAI

v1.0.0Coding Agentsstable

High-agency PowerShell AI framework for multi-agent orchestration and autonomous systems engineering

agentic-aimcpmulti-agent-orchestrationpowershell-coretdd
Share:
264
Stars
0
Downloads
0
Weekly
0/5

What is PSAI?

PSAI is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to high-agency powershell ai framework for multi-agent orchestration and autonomous systems engineering

High-agency PowerShell AI framework for multi-agent orchestration and autonomous systems engineering

This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • High-agency PowerShell AI framework for multi-agent orchestr

Use Cases

Build high-agency multi-agent systems in PowerShell.
Orchestrate autonomous AI workflows at scale.
Implement test-driven development for AI agents.
dfinke

Maintainer

LicenseMIT
Languagepowershell
Versionv1.0.0
UpdatedMay 20, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx psai

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use PSAI

PSAI is a high-agency PowerShell AI framework for building multi-agent systems and autonomous workflows directly in the PowerShell environment. It wraps OpenAI-compatible LLMs with tools like web search (Tavily), stock data, and custom function registration, and supports multi-agent orchestration with test-driven development patterns. Systems engineers and enterprise automation teams use it to embed AI reasoning into PowerShell scripts, runbooks, and DevOps pipelines without leaving the PowerShell ecosystem.

Prerequisites

  • PowerShell 7+ (PowerShell Core) on Windows, macOS, or Linux
  • An OpenAI API key stored as $env:OpenAIKey, or Azure OpenAI credentials
  • Optional: Tavily API key ($env:TAVILY_API_KEY) for web search capability
  • An MCP client such as Claude Desktop if using PSAI as an MCP server
1

Install the PSAI module from the PowerShell Gallery

Install the PSAI module using the standard PowerShell module installer. This works on Windows, macOS, and Linux with PowerShell 7+.

Install-Module -Name PSAI
2

Set your API key

Export your OpenAI API key as an environment variable. Add this to your PowerShell profile ($PROFILE) for persistence.

$env:OpenAIKey = 'sk-your-openai-key-here'
# Optional: for web search
$env:TAVILY_API_KEY = 'tvly-your-tavily-key-here'
3

Import the module and run a basic chat

Import PSAI and make a simple chat call to verify the setup is working.

Import-Module PSAI
Invoke-OAIChat 'How do I list all running services in PowerShell?'
4

Create and run an agent

Create a PSAI agent with custom instructions. The agent can be piped to Get-AgentResponse to answer questions.

$agent = New-Agent -Instructions 'You are a DevOps expert. Provide concise PowerShell solutions.'
$agent | Get-AgentResponse 'Write a script to monitor disk usage and alert if above 80%'
5

Add to your MCP client config

To use PSAI as an MCP server from Claude Desktop, add a server entry pointing to the npx launcher.

{
  "mcpServers": {
    "psai": {
      "command": "npx",
      "args": ["psai"],
      "env": {
        "OpenAIKey": "sk-your-openai-key-here"
      }
    }
  }
}

PSAI Examples

Client configuration

Claude Desktop config for PSAI with the OpenAI key passed as an environment variable.

{
  "mcpServers": {
    "psai": {
      "command": "npx",
      "args": ["psai"],
      "env": {
        "OpenAIKey": "sk-your-openai-key-here",
        "TAVILY_API_KEY": "tvly-your-tavily-key-here"
      }
    }
  }
}

Prompts to try

Example PowerShell and agent prompts using PSAI capabilities.

- "Create a PSAI agent that monitors Windows Event Logs for errors and summarises them hourly."
- "Use the TavilyAITool to search for the latest PowerShell 7.5 release notes."
- "Write a multi-agent pipeline: one agent plans a deployment, another agent executes each step."
- "Convert this GitHub repository to an AI prompt using ConvertTo-AIPrompt so I can ask questions about the code."
- "Get the current stock price of MSFT using the StockTickerTool."

Troubleshooting PSAI

Install-Module fails with 'execution policy' error

Run 'Set-ExecutionPolicy -ExecutionPolicy RemoteSigned -Scope CurrentUser' in an elevated PowerShell session, then retry Install-Module -Name PSAI.

Invoke-OAIChat returns an authentication error

Confirm $env:OpenAIKey is set to a valid OpenAI API key in the current PowerShell session. Use '$env:OpenAIKey' (with the dollar sign) to check its value. The variable must be set before importing the module.

TavilyAITool returns no results

Ensure $env:TAVILY_API_KEY is set to a valid Tavily API key. Sign up at https://tavily.com to get a free key. Without it, the web search tool will fail silently or throw an API error.

Frequently Asked Questions about PSAI

What is PSAI?

PSAI is a Model Context Protocol (MCP) server that high-agency powershell ai framework for multi-agent orchestration and autonomous systems engineering It connects AI assistants to external tools and data sources through a standardized interface.

How do I install PSAI?

Follow the installation instructions on the PSAI GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with PSAI?

PSAI works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is PSAI free to use?

Yes, PSAI is open source and available under the MIT license. You can use it freely in both personal and commercial projects.

Browse More Coding Agents MCP Servers

Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "psai": { "command": "npx", "args": ["-y", "psai"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use PSAI?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides