Metadata-Version: 2.4
Name: prompt-guardrail
Version: 1.0.0
Summary: A fine-tuned ML classification model that detects and blocks jailbreak attempts and prompt injection threats in LLM-based conversational systems.
Author-email: Manikandan R <rmanikandanrjsm@gmail.com>
License-Expression: Apache-2.0
Keywords: guardrails,ml,prompt,safety,llm,jailbreak
Requires-Python: >=3.9
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: scikit-learn
Requires-Dist: numpy
Requires-Dist: joblib
Dynamic: license-file

# Guardrail Training — Input Safety Classifier

pip install pyspark==4.1.1 delta-spark==4.1.0 pandas==2.3.3 requests==2.25.1 python-dotenv==1.2.2 huggingface_hub==1.16.1 fsspec==2026.4.0 fastparquet==2026.5.0
