> ## Documentation Index
> Fetch the complete documentation index at: https://docs.valkyrie.vals.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Quickstart

> Install Valkyrie, configure credentials, and start a benchmark run.

## Prerequisites

* Python 3.12
* [uv](https://docs.astral.sh/uv/) and Git, used by the install command below
* An S3 bucket for benchmark artifacts and agents
* AWS credentials with the [required permissions](/self-hosting/aws-permissions)
* A configured [sandbox provider](/get-started/sandbox-providers)
* A Vals API key when using hosted mode

## Install the CLI

```bash theme={null}
uv tool install git+https://github.com/vals-ai/Valkyrie@prod
```

Both `valkyrie` and `valk` invoke the same CLI.

## Configure Valkyrie

```bash theme={null}
valkyrie config init
```

Choose hosted or self-hosted mode when prompted. See [Choose a hosting mode](/get-started/hosting) for the credentials each mode requires.

## Upload an agent

Push a local agent directory that contains a valid `contract.yaml`:

```bash theme={null}
valkyrie agent push ./agents/sweagent --name sweagent
```

See [Agent contract](/agents/agent-contract) to create a compatible agent.

## Start your first run

```bash theme={null}
valkyrie run start --agent sweagent --benchmark swebench --connect
```

`--connect` streams progress until the run finishes, so you can watch the first run end to end. The command prints the run id; use it with any other `run` command. Add `--model` when the agent contract requires one.

## Next steps

* [Start a run](/runs/start) for run scope, secrets, and repeated runs
* [Monitor runs](/runs/monitor) for streaming, status, and history
* [Results and outputs](/runs/results-and-outputs) for scores and agent files
* [`valkyrie run start` reference](/reference/cli/run#start) for every option and default