# How to watch an agent improve its own setup in 5 minutes, no API key with harness2

Repo: [google-research/harness2](https://github.com/google-research/harness2) | Difficulty: Beginner | Time: 15 minutes | Basis: Seen in the wild

Run the built-in demo to see both recursion modes edit a harness with real Python executions and zero credentials.

**Who it is for:** Curious developer who saw Harness2 trending and wants proof it runs before investing in setup

**The problem:** You cannot tell whether a self-improving harness framework actually works without wiring up models and paying for API calls.

**The outcome:** A runs/demo folder showing parallel and sequential recursion with deterministic editors, produced locally in seconds.

## Steps

1. Create the environment: uv venv --python 3.12 .venv
2. Activate it: source .venv/bin/activate
3. Install the package: uv pip install -e '.[render]'
4. Run the demo: harness2-demo --output runs/demo --k 2
5. Open the runs/demo folder and read the adopted harnesses and reflection notes.

## Prompt to paste into your agent

```text
Clone google-research/harness2, set up a Python 3.12 uv environment, install it, and run harness2-demo with --k 2 into runs/demo. Summarize what files were produced and verify the demo ran without any model credentials.
```

## Caveat

The demo uses deterministic editors, not real models, so it shows the mechanics of the loop and says nothing about real-world score gains.

## Seen in the wild

- X (Oct 2026): https://x.com/i/web/status/2112227371847729175

Dual-use notice: only use this on software, sites and systems you own or are authorized to test.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/google-research-harness2/try-self-improving-agent-demo-no-keys/