# OpenCLI Agent Pack: Ollama

Use this when an AI agent needs to work with `ollama`.

## What this CLI is for
Local models, ai prototyping, and private inference from the terminal. Built by Ollama. Start with `ollama pull llama3.2` and go from there. Runs entirely on your machine.

Best for: local models, ai prototyping, and private inference from the terminal.

## Agent readiness
Usable with guardrails (45/100)
- Supports non-interactive/scripted use.
- Can run locally with less credential exposure.

## Install
```sh
brew install ollama
```

## Verify before real work
```sh
ollama --version
```
Expected signal: Responds locally and is ready for models.

## Safe starting commands
```sh
ollama --version
```

```sh
ollama run llama3.2
```

```sh
ollama pull llama3.2
```

```sh
ollama serve
```

## Guardrails for agents
- Prefer small commands and ask the agent to summarize plain text output.
- Start read-only, then ask before mutations.

## Suggested agent instruction
You may use Ollama (`ollama`) for local models, ai prototyping, and private inference from the terminal.. First install it if missing, then run the verify command. Start with read-only or inspection commands. Summarize what you found before changing anything. Ask for confirmation before commands that mutate remote state, spend money, deploy, delete data, merge code, or expose secrets.

Source: OpenCLI