File size: 1,323 Bytes
13a23cd
 
46beb99
13a23cd
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
4bb08ed
46beb99
4bb08ed
13a23cd
aa08f9f
 
 
9a45547
aa08f9f
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
---
license: cc0-1.0
doi: 10.5281/zenodo.23076190
pretty_name: "UDI Test Data"
tags:
  - test-data
  - synthetic
  - udi
configs:
  - config_name: default
    data_files:
      - split: train
        path: "udi-small.csv"
---

# UDI Test Data

Device identifier, lot, and serial strings in the GS1 element-string shape used by UDI carriers. Synthetic devices; not real products.

Free to use under **CC0-1.0** — public domain dedication, no attribution required.

## Files

- `udi-small.json` / `udi-small.csv` — 100 rows (documents: 25)
- `udi-medium.json` / `udi-medium.csv` — 2,000 rows (documents: 250)
- `udi-large.json` / `udi-large.csv` — 20,000 rows (documents: 2,500)

Deterministic: the same size always produces the same rows.

## Source

Generated by StanzaAPI from <https://stanzaapi.com/datasets/udi>.

## Citation

```
StanzaAPI. (2026). UDI Test Data. CC0-1.0.
DOI: 10.5281/zenodo.23076190
https://stanzaapi.com/datasets/udi
```

## Security: adversarial corpus

This dataset ships a catalogued **adversarial corpus** of 11 hostile inputs
(injection, XXE, resource exhaustion, encoding attacks, validation bypass, structural confusion,
and prompt injection for AI agents) with the safe expected behaviour. See
[`adversarial/`](adversarial/) or the [hub](https://stanzaapi.com/datasets/adversarial).