> Data revision: baf7c2f4307d3a989a8fb9363df8000fda3d9310
> Source files:
- [reproduce/lab-shaping-baseline-noise2-20261108/README.md](https://github.com/KogenAI/kogen-bench/blob/baf7c2f4307d3a989a8fb9363df8000fda3d9310/reproduce/lab-shaping-baseline-noise2-20261108/README.md)
- [categories/harness-levers.md](https://github.com/KogenAI/kogen-bench/blob/baf7c2f4307d3a989a8fb9363df8000fda3d9310/categories/harness-levers.md)
- [site/content/finding-pages.json](https://github.com/KogenAI/kogen-bench/blob/baf7c2f4307d3a989a8fb9363df8000fda3d9310/site/content/finding-pages.json)

# Variation in shaping readiness

- **Finding id:** `lab-shaping-baseline-noise2-20261108`
- **Verdict:** Unresolved
- **Finished:** 8 November 2026
- **Comparison:** [Kogen shaping component](/comparisons/kogen-shaping/)
- **Reproduction pack:** [reproduce/lab-shaping-baseline-noise2-20261108/](https://github.com/KogenAI/kogen-bench/tree/baf7c2f4307d3a989a8fb9363df8000fda3d9310/reproduce/lab-shaping-baseline-noise2-20261108) (data, analysis script and rerun steps)
- **Category page:** [categories/harness-levers.md](https://github.com/KogenAI/kogen-bench/blob/baf7c2f4307d3a989a8fb9363df8000fda3d9310/categories/harness-levers.md)

## In plain terms

An unpatched repeat reached 7 of 10, with 2 gains and 1 losses versus the stored run.

Verdict key: **Yes**, The result supports the change or claim that was tested. **No**, The result does not support it, so nothing changes. **Unresolved**, The evidence is not strong enough either way. **Superseded**, A later test replaced this result.

## Full text of the reproduction pack

The text below is the pack's README as published in the repository (a system folder name may show without its leading slash, so that the site's privacy scan passes).

On ten small requests, the stored release and an unpatched repeat each ran once with Sol at low effort and no answerer. The stored result was 6/10 ready; the repeat was 7/10. 2 cases changed from not ready to ready and 1 from ready to not ready. This measures run-to-run variation, not a candidate improvement. Intermediate refusals could be repaired; readiness requires no terminal refusal or question.

The elapsed times include failed attempts. The shared credit balance showed no movement during these launches; exact attributable spend is unknown. The small Go and Node workspaces are reconstructions whose byte identity with the original study is unverified. No held-out or complex requests, Build stage, or hidden checks were run. Landing and hidden-check outcomes are unmeasured.

Both runs used [the same release source](https://github.com/KogenAI/kogen/tree/47236bbb93385bc960fd2ac353c9b000bd6ed571) and [shaping source](https://github.com/KogenAI/kogen-shaping/tree/0484edf817f6dc74f691f44d5234cf7f6ae0435e). One paired repetition is too small to characterize the full distribution of variation.

## Rerun

Run `python3 report.py` to recompute the paired counts from the data. To repeat the model study, follow the [small-request setup](/findings/lab-shaping-small-requests-20261107/), use the pinned release and the same ten requests, model and low effort with no answerer, and record terminal readiness, questions and elapsed time.
