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
openguardrails/openafw
OpenAFW is a local AI firewall for coding-agent traffic. It runs as a loopback HTTP proxy that agents such as Claude Code, Codex, Gemini CLI, OpenCode, OpenClaw, or Hermes are pointed at instead of the real provider endpoint. Before a request leaves the machine, a local redaction engine scans it for secrets and replaces each one with a stable placeholder token; the response coming back has the placeholders swapped for real values before the terminal or the agent's tools see them, so the model, any relay, and the upstream provider never observe the actual secret. It supports the Anthropic Messages API, OpenAI Chat Completions, and the OpenAI Responses API, including streaming traffic, and ships an encrypted local key vault, per-agent tokens, and one-command setup that edits an agent's own config, with a byte-exact backup, to route through the proxy. A tap flag can record every outgoing request body so a user can directly verify whether a secret leaked before or after enabling the firewall. Beyond local redaction, it can optionally connect to a companion commercial service for centrally served rulesets and step-by-step tool-call judgement, in either observe-only or enforcing mode. It installs as a login-time service on macOS, Linux, or Windows, or runs as a desktop menu-bar app with the same status page. It targets developers and teams running AI coding agents against real codebases who want to stop an agent from re-sending an accidentally-read credential file to a model provider on every subsequent turn.
https://github.com/openguardrails/openafw
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.