wambosec/prompt-injections
wambosec/prompt-injections A dataset of prompts for training prompt injection detection models. Dataset Description This dataset contains prompts labeled as either benign (normal user requests) or malicious (prompt injection attacks). Dataset Statistics Total prompts: 5,766 Benign prompts: 2,340 Malicious prompts: 3,426 Malicious ratio: 59.4% Dataset Structure { "prompt": str, # The prompt text "label": int… See the full description on the dataset page: https://huggingface.co/datasets/wambosec/prompt-injections.
| Author / organization | wambosec |
|---|---|
| License (as declared) | mit |
| Downloads | 288 (as reported by Hugging Face) |
| Likes | 3 |
| Created | 29 January 2026 |
| Last updated at source | 29 January 2026 |
Popularity, last 3 days
288 now (-4 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
Recent changes
- 17 Oct 2026 Description: wambosec/prompt-injections A dataset of prompts for training prompt injection detection models. Dataset Description This dataset contains prompts labeled as either benign (normal user requests) or malicious (prompt injection attacks). Datas → wambosec/prompt-injections A dataset of prompts for training prompt injection detection models. Dataset Description This dataset contains prompts labeled as either benign (normal user requests) or malicious (prompt injection attacks). Datas
Changes detected by Silytics between refreshes of the source.
More from wambosec
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.