| --- |
| library_name: transformers |
| pipeline_tag: text-classification |
| license: other |
| license_name: skillscreen-proprietary |
| license_link: LICENSE |
| base_model: AlexWortega/openjev |
| datasets: |
| - ProtectSkills/MaliciousSkillBench |
| tags: |
| - skillscreen |
| - agent-security |
| - nli |
| - sd |
| --- |
| |
| # SkillScreen-2B-SD |
|
|
| This model is described in the paper: [SkillScreen: High-Recall Screening of Malicious Agent Skills with Decision Models](https://huggingface.co/papers/2610.17727). |
|
|
| 2B decision model for detecting malicious agent skills from `SKILL.md`. Trained on MaliciousSkillBench using Source-Disjoint (SD), with an 8K input budget. |
|
|
| Based on [OpenJev 2B](https://huggingface.co/AlexWortega/openjev). Full weights are at the root. The original LoRA adapter and classification head are in `adapter/`. |
|
|
| ## Use |
|
|
| Requires CUDA with BF16 support. |
|
|
| ```bash |
| pip install huggingface_hub |
| hf download SkillScreen/SkillScreen-2B-SD --revision v1 --local-dir skillscreen |
| pip install -r skillscreen/requirements.txt |
| python skillscreen/score.py path/to/SKILL.md |
| ``` |
|
|
| Native maliciousness score **≥ 0.5** is flagged as suspicious. |
|
|
| [Checkpoint details](provenance.json) · [Export verification](merge_verification.json) · [Project page](https://jerroldh.github.io/SkillScreen-webpage/) · [Code](https://github.com/JerroldH/SkillScreen) |
|
|
| Jerrold Huang, Junwei Quan, Saiyue Lyu, Zihan Guo, Yingqiang Ge. |
|
|
| [License](LICENSE) · [Third-party notices](THIRD_PARTY_NOTICES.md) |
|
|
