AI on DecipherMiddleware

CLI Coding Agents Meet MuleSoft MCP Server

Wed, 10 Jun 2026 00:00:00 +0530

As an Integration and DevOps engineer, I’ve spent countless hours staring at black and white screens — vim, vi, nano, writing, running, testing, and shipping scripts, all from one home:

The OG

echo "terminal"

Coding agents just moved in.

## Antigravity
agy
## GitHub Copilot
copilot

After experimenting with IDE coding agents — building MuleSoft applications using MuleSoft Vibes/ACB right inside VS Code — I wanted to take it further. Back to where I’m most comfortable.

Let’s spin up a CLI coding agent, and arm it with MuleSoft DX MCP Server to interact with Anypoint Platform - straight from the terminal.

MuleSoft DX MCP Server
- Features
Prerequisites
Connected App Creation
The OG: Terminal
CLI Coding Agents
- Antigravity
  - MCP Server configuration
  - Moment of truth
- Copilot
  - MCP Server configuration
  - Moment of truth

MuleSoft DX MCP Server

MuleSoft DX MCP Server supports API design, application deployment, and asset management

Features

API specification development
Application creation and configuration
Agent network creation
Asset publication in Exchange
Instance deployment and management to CloudHub 2.0 and Runtime Fabric
Application security

These features look very promising for a CLI agent. Let’s set them up.

Prerequisites

Enable Agentforce for Anypoint platform - Ref
Install Node.js 20 or later - Ref
Organization admin access to create a connected app - Ref

Connected App Creation

Log in to Anypoint Platform.
Under Access Management, click Connected Apps.
Click Create App. Add a name for the connected app.
Select App acts on its own behalf (client credentials).
Click on Add Scopes. Add the scopes mentioned in the below table.

Anypoint Code Builder	Mule Developer Generative AI User
Anypoint Monitoring	Monitoring Viewer
Anypoint Code Builder	Mule Developer Generative AI User
API Manager	Manage APIs Configuration
	Manage APIs Configuration
	View APIs Configuration
Exchange	Exchange Administrator
	Exchange Contributor
	Exchange Creator
	Exchange Viewer
General	View Connected Applications
General	View Organization
OpenID	Profile
Runtime Manager	Cloudhub Network Viewer
	Create Applications
	Read Applications
	Read Runtime Fabrics

Click on Save to save the new connected app.

The OG: Terminal

Now open The OG: Terminal. Set the environment variables for the terminal session.

macOS/Linux

export ANYPOINT_CLIENT_ID="<CLIENT_ID>"
export ANYPOINT_CLIENT_SECRET="<CLIENT_SECRET>"

Optional

export ANYPOINT_REGION="<REGION_NAME>"

Windows

$env:ANYPOINT_CLIENT_ID="<CLIENT_ID>"
$env:ANYPOINT_CLIENT_SECRET="<CLIENT_SECRET>"

Optional

$env:ANYPOINT_REGION="<REGION_NAME>"

CLI Coding Agents

Setup is complete on the Anypoint side and the env variables are set in the terminal. Let’s make the terminal smarter and equip it with MuleSoft DX MCP Server.

Antigravity
Copilot

Antigravity

Install Antigravity CLI following the link
Run the below command to start CLI. Log in using your google/gemini account.

agy

Type /mcp, press Enter.

MCP Servers

 No MCP servers configured.

Keyboard: ↑/↓ Navigate enter Actions

esc to cancel

As of now, no MCP servers are configured for antigravity-cli.

MCP Server configuration

Antigravity lets you setup MCP server as Global setup or Workspace Local setup.

Global: ~/.gemini/antigravity-cli/mcp_config.json

Workspace local setups: .agents/mcp_config.json

Update mcp_config.json as below.

{
 "mcpServers": {
 "mulesoft": {
 "command": "npx",
 "args": ["-y", "mulesoft-mcp-server", "start"]
 }
 }
}

Make sure that env variables were properly set for the terminal session.
Let’s start antigravity.

agy

Type /mcp, press Enter.

MCP Servers
Plugins (~/.gemini/antigravity-cli/plugins)
> ✓ mulesoft Tools: create_mule_project, generate_mule_flow, create_mcp_server, deploy_mule_application, list_applications, +16 more
Keyboard: ↑/↓ Navigate enter Actions

If you see the message similar to above then you are all set and ready to perform operations 😄

But wait, is it too easy????

Let’s get the hands dirty 😈

My initial prompts didn’t work and, it was not able to list the MCP tools 😞 Countless hours and tokens wasted 😫!!!

I kept on trying…. I kept trying, and then I changed the /model to Claude Sonnet 4.6, and prompted:

list mcp tools

A good output 😎

Another prompt

List deployed applications in dev environment.

Another success 😎

Copilot

Install the copilot following the link
Run the below command to check if copilot is working.

copilot

Type /login and press enter. Follow instructions to log in to your GitHub account and activate GitHub Copilot.
To view list of available MCP servers type /mcp show and press enter.

As of now only GitHub MCP server is enabled.

MCP Server configuration

Type /mcp add and press enter. Fill in details as per below table.

Server Name:	mulesoft-dx
Server Type:	STDIO
Command to start the server:	npx -y mulesoft-mcp-server start
Environment variables:	Add the env variables of ANYPOINT_CLIENT_ID and ANYPOINT_CLIENT_SECRET if not configured in terminal env else keep it blank.
Tools:	*

ctrl+s to save the MCP server configuration.

Moment of truth

Prompt

List deployed applications in dev environment.

Now, CLI agent has got MuleSoft DX MCP Server.

References

Installing AI models Localhost

Mon, 16 Feb 2026 00:00:00 +0000

Everything works on my Localhost

One day, I started wondering, is it possible to run AI models on a normal system where I play games? I recalled that I have a NVDIA graphics card in my desktop. I started digging into what I can do. I got so many doubts. Will the desktop be able to support? If yes, where do I start? Is it going to crash the system?

Generated By AI (ChatGPT)

So many questions. Let’s ask gen AI (ChatGPT/Gemini…etc).

On putting the query. It started giving me answers that need huge servers and configurations. No! No!…. !!! But do I have that much computing power? I need the answers that fit my existing computing resources.

First things first, I grabbed a piece of paper. Oh wait! I know this works, but in today’s digital world. VsCode becomes the new notepad :P. I gathered a couple of pieces of information.

Desktop Configurations

Component	Configurations
CPU	i5 11400 11th Generation
RAM	32 GB
SSD	Yes
GPU	MSI GeForce GTX 1650 VENTUS XS OC Nvidia Graphic Card
OS	Windows 11

Configurations, as per me looks decent. Will it be able to support any LMs? Answer is YES. Let’s give it a try.

Pre-requisites

1. Update Windows

Update operating system. I am using Windows, and performed a Windows update.

2. Update GPU Drivers

Update GPU drivers. I have an NVidia GPU, so updating the NVidia GPU drivers.
Enable the CUDA cores. Download the CUDA toolkit.
https://developer.nvidia.com/cuda/toolkit
If the GPU is older, like mine, use the link below to find the compute compatibility version for the CUDA toolkit. https://developer.nvidia.com/cuda/gpus
For the GTX 1650, it supports version 7.5.
https://developer.nvidia.com/cuda-75-downloads-archive

Run the command in cmd or PowerShell below to check if the drivers and the CUDA toolkit are installed.

nvidia-smi

Output will look like this.

+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 591.86 Driver Version: 591.86 CUDA Version: 13.1 |
+-----------------------------------------+------------------------+----------------------+
| GPU Name Driver-Model | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+========================+======================|
| 0 NVIDIA GeForce GTX 1650 WDDM | 00000000:01:00.0 On | N/A |
| 40% 36C P8 15W / 90W | 681MiB / 4096MiB | 4% Default |
| | | N/A |
+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+
| Processes: |
| GPU GI  CI PID Type  Process name GPU Memory |
| ID ID Usage |
|=========================================================================================|
| 0 N/A N/A 1816 C+G ...8bbwe\PhoneExperienceHost.exe N/A |
| |
+-----------------------------------------------------------------------------------------+

3. WSL Installation

Since I am running Windows, I need a Linux distribution to run an LM locally. To do so, I have options such as Docker, a virtual machine or WSL. To take advantage of GPUs in a virtual environment, WSL is the best option.

Check current WSL distros installed on the system.

wsl --list --verbose

Install Ubuntu 22.04 using WSL 2.

wsl --install -d Ubuntu-22.04

After downloading the distros, it will ask for username and password. Provide username and password,which will be used to login.

Check installation

wsl --list --verbose
 NAME STATE VERSION
* Ubuntu-22.04 Stopped 2

Moving installation to other directory

Export the installation to tar, to move the installation to another directory.

wsl --export Ubuntu-22.04 D:\AI\ubuntu.tar

Unregister the current installation from the list of installed distros.

wsl --unregister Ubuntu-22.04

Import to another location.

wsl --import Ubuntu-22.04 D:\AI\ubuntu D:\AI\ubuntu.tar --version 2

wsl -d Ubuntu-22.04

Update the default user.

echo -e "[user]\ndefault=your_username" | sudo tee /etc/wsl.conf

Restart WSL.

wsl --shutdown

wsl -d Ubuntu-22.04

Check if drivers are installed correctly inside WSL/Ubuntu. Run the below command inside the same window.

nvidia-smi

Okay,! Now the desktop is ready to install and run LMs. But which one to run? How to interact with LM? How to manage LMs locally?

Install Ollama

Let’s use Ollama 😎. Docs: https://docs.ollama.com/

Run below command to install Ollama.

curl -fsSL https://ollama.com/install.sh | sh

Verify installation.

ollama --version

Create a directory to save ollama models. Create an environment variable to point to the new location.

mkdir <path-to-store-ollama-models>
nano ~/.bashrc
export OLLAMA_MODELS=<path-to-store-ollama-models>
source ~/.bashrc
## Validate if environment variable is set properly
echo $OLLAMA_MODELS

Start Ollama

ollama serve

Now Ollama will start running. In another terminal, verify that Ollama is running:

ollama -v

ollama version is 0.15.2

I am done with the hard part 😎😎😎.

Now, let’s download the AI model. To be on the safe side, I started small by downloading a vector embedding model (nomic-embed-text).

ollama pull nomic-embed-text

Test the downloaded model.

curl --location 'http://localhost:11434/api/embeddings' \
--header 'Content-Type: application/json' \
--data '{
 "model": "nomic-embed-text",
 "prompt": "deciphermiddleware"
 }'

A vector output is generated. A successful test!!!

How much LM work I can offload to the GPU depends on VRAM a lot. Since the GTX 1650 has only 4GB VRAM, it will not allow large models to run on the GPU. Thus, models will run on a shared basis between CPU and GPU. Let me try a 3b parameter model llama3.2.

ollama pull llama3.2
ollama run llama3.2

Output

>>> hi
How can I assist you today?

DEBUG INFO

load_tensors: loading model tensors, this can take a while... (mmap = true)
load_tensors: offloading 26 repeating layers to GPU
load_tensors: offloaded 26/29 layers to GPU
load_tensors: CPU_Mapped model buffer size = 1918.35 MiB
load_tensors: CUDA0 model buffer size = 1488.14 MiB
llama_context: constructing llama_context
llama_context: n_seq_max = 1
llama_context: n_ctx = 4096
llama_context: n_ctx_seq = 4096
llama_context: n_batch = 512
llama_context: n_ubatch = 512
llama_context: causal_attn = 1
llama_context: flash_attn = auto
llama_context: kv_unified = false
llama_context: freq_base = 500000.0
llama_context: freq_scale = 1
llama_context: n_ctx_seq (4096) < n_ctx_train (131072) -- the full capacity of the model will not be utilized
llama_context: CPU output buffer size = 0.50 MiB
llama_kv_cache: CPU KV buffer size = 32.00 MiB
llama_kv_cache: CUDA0 KV buffer size = 416.00 MiB
llama_kv_cache: size = 448.00 MiB ( 4096 cells, 28 layers, 1/1 seqs), K (f16): 224.00 MiB, V (f16): 224.00 MiB
llama_context: Flash Attention was auto, set to enabled
llama_context: CUDA0 compute buffer size = 588.73 MiB
llama_context: CUDA_Host compute buffer size = 14.01 MiB
llama_context: graph nodes = 875
llama_context: graph splits = 29 (with bs=512), 3 (with bs=1)

Now, let’s start using the AI models and explore them more. But that will be for some other day.

I hope you like the journey. Please share your valuable feedback. 😊😊😊

AI on DecipherMiddleware

CLI Coding Agents Meet MuleSoft MCP Server

The OG

Table of Contents

MuleSoft DX MCP Server

Features

Prerequisites

Connected App Creation

The OG: Terminal

macOS/Linux

Optional

Windows

Optional

CLI Coding Agents

Antigravity

MCP Server configuration

Let’s get the hands dirty 😈

Copilot

MCP Server configuration

Moment of truth

Prompt

References

Installing AI models Localhost

Desktop Configurations

Pre-requisites

1. Update Windows

2. Update GPU Drivers

3. WSL Installation

Install Ollama