Context Hunter

Set up

To set up the repository run:

./setup.sh

Then configure one of the supported LLM backends:

• Groq API — runs models remotely using the Groq API
• Ollama — runs models on the local device

Groq API

Get Groq API key and set the key as the local variable before running the application:

read -s GROQ_API_KEY
export GROQ_API_KEY

Then in the application config use groq: prefix, eg. "groq:llama3.1"

Ollama

Install and run Ollama, then download the model you want to use:

ollama pull gemma3:12b

Then in the application config use ollama: prefix, eg. "ollama:gemma3:12b"