# LeRobot Diffusion Policy PushT closed-loop evaluation: robotics skill example

LeRobot Diffusion Policy PushT closed-loop evaluation. Source-pinned, bounded simulation with explicit goal checks; no physical execution.

## Task and environment

lerobot | 0.3.3 | pusht-2d-pusher | diffusion_policy | linux | 3.10

## Example input

```json
{
  "seed": 0
}
```

## Retrieve and prepare

Read https://hub.superfastrobotics.com/skills/lerobot-diffusion-pusht/skill.json or https://api.superfastrobotics.com/v1/skills/lerobot-diffusion-pusht?hub_commit=ede75c00118e9808e5099da43c26f4619cc3bc37&record_version=1.0.0&include=experience without installing a client. Inspect the license, exact environment and evidence before preparation. Preparation and execution remain separate.

Source: https://github.com/huggingface/lerobot at b883328e6c95681ca90a18b102e4ae5e1f91e2bf. Record version: 1.0.0.

## Success condition

All declared goal and finite-state checks pass; exit 0 with task_complete. Original PushT success flag and measured object/target overlap >=95%; no accuracy claim beyond the three retained seed episodes. Checkpoint license: Apache-2.0, preserved at its immutable model revision.

## Recorded evidence

import_verified: 1/3 frozen cases completed; retained failures are not promoted.

This page is a data-only example and performs no execution.