ProtectSkills/MaliciousSkillBench
Viewer • Updated • 65.6k • 791 • 2
How to use SkillScreen/SkillScreen-2B-Package-Experimental with Transformers:
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("text-classification", model="SkillScreen/SkillScreen-2B-Package-Experimental") # pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForSequenceClassification
processor = AutoProcessor.from_pretrained("SkillScreen/SkillScreen-2B-Package-Experimental")
model = AutoModelForSequenceClassification.from_pretrained("SkillScreen/SkillScreen-2B-Package-Experimental", device_map="auto")Experimental 2B decision model for SKILL.md and related file contents. Trained on the recovered SRC010 package cohort with a 32K input budget.
Based on OpenJev 2B. Full weights are at the root. The original LoRA adapter and classification head are in adapter/.
Requires CUDA with BF16 support.
pip install huggingface_hub
hf download SkillScreen/SkillScreen-2B-Package-Experimental --revision v1 --local-dir skillscreen
pip install -r skillscreen/requirements.txt
python skillscreen/score.py path/to/package.json
Native maliciousness score ≥ 0.5 is flagged as suspicious.
package.json contains Markdown and companion file text.
{"skill_text": "Contents of SKILL.md", "files": [{"path": "scripts/helper.py", "text": "File contents"}]}
Checkpoint details · Export verification · Project page · Code
Jerrold Huang, Junwei Quan, Saiyue Lyu, Zihan Guo, Yingqiang Ge.