Microsoft's open-source PII detection and anonymization engine supporting 50+ entity types across text, images, and structured data.

MIT-licensed security toolkit by ProtectAI that sanitizes LLM prompts and responses — blocking prompt injection, toxic content, PII leakage, and secrets.

NVIDIA's open-source toolkit for adding programmable safety, topical, and quality guardrails to LLM-based conversational systems.

Structured outputs for LLMs using Pydantic — patches any OpenAI-compatible client to return validated Python objects.

Guaranteed structured generation from LLMs: enforce JSON, regex, grammar, or Pydantic schemas at the token level.

Validate, fix, and filter LLM outputs — define structured schemas and safety rules, then automatically retry when outputs fail validation.

ProtectAI's self-hardening prompt-injection detector using a multi-stage defence: heuristics, LLM analysis, and a vector canary database.

Red-teaming kit for agentic LLM workflows: 2000+ adversarial probes, RL-crafted injections, CI-ready.

ProtectAI's scanner that detects malicious payloads hidden inside pickle, PyTorch, TF, and Keras model files.