Announcing EnforceCore v1.0

We are thrilled to announce the general availability of EnforceCore v1.0, the first runtime enforcement layer designed specifically for AI agents.

Why Runtime Enforcement?

As AI agents become more autonomous, the risks associated with their actions increase. Static analysis and prompt engineering are no longer sufficient to guarantee safety. Agents need a runtime guardrail that enforces policies at the execution level.

EnforceCore provides:

• Deterministic Policy Enforcement: Define what your agent can and cannot do in YAML.
• PII Redaction: Automatically mask sensitive data before it leaves your environment.
• Audit Logging: Keep a tamper-proof record of every action your agent takes.

What's New in v1.0?

1. Stable Policy Engine

The policy engine has been rewritten in Rust for performance and reliability. It now supports complex logic, including conditional allowances based on budget, time, and resource usage.

2. Zero-Dependency Python Adapter

We've released a new Python adapter that works with any agent framework, including LangChain, CrewAI, and AutoGen. It requires no external dependencies and adds negligible overhead.

3. Compliance Ready

EnforceCore now includes built-in mappings for the EU AI Act and NIST AI RMF. You can generate compliance reports directly from your audit logs.

Getting Started

To get started with EnforceCore, check out our Quickstart Guide or install it directly via pip:

pip install enforcecore

We can't wait to see what you build with EnforceCore!