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
praetorian-inc/julius
Julius is an LLM service fingerprinting tool aimed at security professionals doing penetration tests, attack-surface discovery and infrastructure assessments. It answers a narrow question quickly: given a reachable endpoint, which AI serving software is behind it? That is distinct from model fingerprinting, which tries to identify the model that produced a piece of text — Julius identifies the server, so an open port becomes a known service such as Ollama, vLLM, LiteLLM, LocalAI, Hugging Face TGI or an AWS Bedrock proxy rather than an unlabelled HTTP listener. It works by sending targeted HTTP probes and matching the responses against signatures for 63 services, prioritizing probes by the default ports each service is known to use and running them concurrently so a scan finishes in seconds. Matches are ranked by a one-to-hundred specificity score, so a precise match such as LiteLLM outranks a generic OpenAI-compatible response, and where the identified service exposes a model list Julius extracts the available models as part of the result. Inputs may be a single target, a file of targets or a stream on stdin, and output can be a table for humans or JSON and JSONL for pipelines. New detections are contributed as simple YAML probe files rather than code. The tool operates entirely offline with no cloud dependency, compiles to a single portable Go binary, and installs with a single go install command. It is published by Praetorian, is open source, and fits into the discovery stage of an AI infrastructure security review or an internal inventory of who is running what.
https://github.com/praetorian-inc/julius
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.