Skip to content

YecoAI-Security-Layer v 1.0

Latest
Latest

Choose a tag to compare

View all tags
@MarcoNasi MarcoNasi released this 25 Apr 23:19
· 1 commit to main since this release
1.0
743bc96

Release v1.0.0 - Initial Professional Release

We are proud to announce the first official release of the YecoAI Security Layer, a deterministic, ultra-low-latency heuristic security filter designed to protect LLM systems from malicious inputs and unsafe outputs.

Key Features

  • 🛡️ RoboticsEngine (Asimov Compliance): Injects foundational safety rules (The Three Laws of Robotics) into the system prompt for native ethical alignment.
  • 🧠 SafetyModel (Advanced Filtering): Context-Aware N-Gram analysis, <think> tag scanning (Chain of Thought), and Enterprise Data Loss Prevention (DLP) for sensitive secrets.
  • 📦 Execution Sandbox Guard: Runtime interception and control of tool usage (Filesystem, Shell, and Network access).
  • 📜 Declarative Policy System: YAML/JSON-based configuration for easy integration into SOC2/GDPR compliant environments.
  • 🌐 Integrations: Out-of-the-box support for LangChain and LlamaIndex.

Installation

Install the stable version directly from PyPI:

pip install yecoai-security-layer

License

This project is dual-licensed:

  • Open Source: Apache 2.0 (for personal, research, and educational use).
  • Commercial: Required for SaaS, enterprise, and paid products. See COMMERCIAL_LICENSE.md for details.

Developed with ❤️ by YecoAI.

Assets 2
Loading