data-privacy-stack/presidio
Presidio is an open-source framework for detecting, redacting, masking, and anonymizing sensitive data (PII) across text, images, and structured data, leveraging NLP and customizable pipelines.
Awesome Infra for AI › AI Safety & Guardrails
hoophq/hoop
Hoop is a single-binary, MIT-licensed sidecar proxy that gives AI agents runtime access controls when they interact with backend systems such as databases. It sits transparently between an agent and the resource it talks to (for example a Postgres connection over the pgwire protocol), so the agent itself needs no SDK, prompt changes, or awareness that the sidecar exists. Two capabilities are highlighted: data masking, which rewrites sensitive fields such as email addresses in the response before it reaches the agent while leaving the underlying request untouched, and policy-based guardrails, which can block destructive operations such as DROP, DELETE, and TRUNCATE by returning a real protocol-level error with a configurable message, so the agent understands it was refused rather than encountering a dropped connection. Configuration is a single YAML file defining masking rules (by entity type and strategy) and policy rules (by operation type), and the software installs as a single binary via Homebrew or Docker. An admin HTTP endpoint exposes health, stats, configuration, and event information. The project frames its purpose as making agents safe to grant access to real data and systems, providing runtime enforcement that does not depend on the agent's own behavior or instructions being followed correctly. It targets teams that want to let coding or data agents query production databases without giving them unrestricted read or write access, or without hand-writing a custom proxy for each protected resource.
https://github.com/hoophq/hoop
Also in AI Safety & Guardrails
NVIDIA-NeMo/Guardrails
NVIDIA NeMo Guardrails is an open-source toolkit for adding programmable guardrails to LLM-based conversational applications, focusing on safety, security, and controlled dialog.
superagent-ai/superagent
Superagent is an open-source SDK providing safety features for AI applications, including prompt injection detection, PII redaction, repository scanning for threats, and red teaming capabilities fo...
Tencent/AI-Infra-Guard
AI-Infra-Guard is a full-stack AI red teaming platform providing comprehensive security analysis, vulnerability scanning, and jailbreak evaluation for AI ecosystems and LLMs.
microsoft/agent-governance-toolkit
AI Agent Governance Toolkit (AGT) provides policy enforcement, identity management, execution sandboxing, and reliability engineering to secure autonomous AI agents in production.
FailproofAI/failproofai
Observability and policy enforcement for AI agent harnesses, hooking twelve coding and chat harnesses to record every run and block dangerous tool calls before they execute.
protectai/llm-guard
LLM Guard is a comprehensive open-source security toolkit designed to fortify Large Language Model (LLM) interactions by providing robust sanitization, malicious content detection, data leakage pre...
lennney/stop-that-shit
Multi-platform hook and skill guard for AI coding agents that blocks unrequested work such as generated hashes, checksums and task-scope expansion at the harness hook boundary.