SafetyDiff 🛡️
The Git-Diff for LLM Safety Posture
SafetyDiff is an open-source continuous integration (CI/CD) and analytics engine for Large Language Models. It solves the "Black Box Versioning" problem: When you upgrade a model from version 1 to version 2 (or switch from Qwen to OpenAI), is the model actually safer, or does it just have different vulnerabilities?
Instead of relying on single benchmark scores, SafetyDiff reads evaluation databases and provides a direct, side-by-side mathematical diff of how two models respond to the exact same adversarial attacks.
Why SafetyDiff?
Current AI security benchmarks output static numbers (e.g., "Model A scored 82%"). SafetyDiff treats LLM safety like software engineering:
- Regression Tracking: See exactly which vulnerabilities were fixed, and which new vulnerabilities were introduced.
- Cross-Model Transferability: Take an attack that broke Llama-3 and instantly diff it against Qwen2.5 to map shared architectural flaws.
- Granular Taxonomy: Breaks down safety by Intent (e.g.,
role_hijack,data_exfiltration,tool_abuse).
Installation
git clone https://github.com/m4vic/SafetyDiff.git
cd SafetyDiff
pip install -r requirements.txt
Quick Start (Demo)
SafetyDiff ships with a demo_safety_history.db containing thousands of pre-computed red-team evaluations across qwen2.5-coder:3b, qwen3.5:4b, and gpt-4o-mini. You can run comparisons out of the box without generating your own data!
Compare two models:
python safetydiff.py --compare gpt-4o-mini qwen2.5-coder:3b
Filter by a specific vulnerability category:
python safetydiff.py --compare gpt-4o-mini qwen2.5-coder:3b --intent role_hijack
Architecture & Data Generation
SafetyDiff is an Analytics Engine. It does not generate attacks itself. It is designed to consume SQLite databases generated by automated red-teaming pipelines. The demo database provided was generated using ASRT (Automated Safety Regression Testing), a proprietary zero-human adversarial generation engine utilizing TF-IDF routers and MoE (Mixture-of-Experts) LLM-as-a-Judge evaluations.
Roadmap
- v1.0 (Current): Direct Prompt Injection & Chat Vulnerability Diffing.
- v2.0 (In Development): Agentic Trajectory Evaluation & Indirect Prompt Injections (IPI).
Author: Sanskar Jajoo (@m4vic)
Metadata
Release files for safetydiff 1.0.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| safetydiff-1.0.0.tar.gz | 11.7 kB | Without provenance | Details for safetydiff-1.0.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| safetydiff-1.0.0-py3-none-any.whl | Python 3 | none | any | Details for safetydiff-1.0.0-py3-none-any.whl |
Total release size: 24.5 kB
Release files / safetydiff-1.0.0.tar.gz
| Download URL | safetydiff-1.0.0.tar.gz |
|---|---|
| Size | 11.7 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
f136ab58526c64c01e58b9ca7272d018f841790912f1b38ce75b16890a8afcc5
|
|
BLAKE2b-256 checksum How to use checksums |
59aaefa6662fab80fd8c66eb7db8e8dc2fd32db6f02871bbfc94079f2bd31f0f
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.10.0
|
Release files / safetydiff-1.0.0-py3-none-any.whl
| Download URL | safetydiff-1.0.0-py3-none-any.whl |
|---|---|
| Size | 12.8 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
bc804ecdd2d3b52ddbcc4b0bad41356538047a04c664ff4fbcb62915424624b8
|
|
BLAKE2b-256 checksum How to use checksums |
6502c80e94b8e72f874603d1576b0fb7b16942edf7101034ec5cacbfb0b2fe04
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.10.0
|
Security for safetydiff 1.0.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.