Post cover image
Kali MCP Server + Gemini AI: Building an AI-Powered Bug Bounty Workflow

June 15, 2026

GEMINI + KALI MCP SERVER FOR BUG BOUNTY

परिवर्तनमेव विकासस्य मार्गः।

Pavanshanmukhmadhav

1 min read

Transliteration: parivartanam eva vikāsasya mārgaḥ

Meaning: "Change itself is the path to growth."

None

Install Node.js using the follow command

sudo apt-get install -y nodejs

Globally install gemini-cli

sudo npm install -g @google/gemini-cli

create a workspace with

mkdir bugbounty_workspace

Create an api key from https://aistudio.google.com/app/apikey

Add it up permanently

echo 'export GEMINI_API_KEY="YOUR_KEY"' >> ~/.zshrc
source ~/.zshrc

Let's check with a prompt

gemini “Hello”
None

other than gemini you can use llama3(No need of Internet),GPT-04,pentestgpt

now further proceed with installation of kali mcp server

sudo apt install mcp-kali-server

configuriation of mcp server

After installation using the kali-server-mcp as it runs the kali Linux Api Server we have to run this before running api server:

kali-server-mcp --port 9199

Now we will run the other executable which is mcp-server. This will make requests to the api .

mcp-server --server http://127.0.0.1:9199

Now we will connect GEMINI with mcp server

gemini-cli mcp add kali-mcp mcp-server --server http://127.0.0.1:9199/

To check if connection is successful

gemini-cli mcp list
None

run the below command in the gemini-cli to check list of avilable tools with the mcp server which are about 12 .

\mcp list
None

GEMINI.md is a configuration file that defines custom behaviors and rules for the Gemini CLI. When you save it to ~/.gemini/GEMINI.md . Gemini reads it and applies your instructions automatically.

Now let's create a recon automation workflow

None

Happy Hunting :>

Found this useful? Share it with your fellow hunters.

Questions or suggestions? Reach out below.

🛡️Hunt responsibly