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
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.
kenryu42/cc-safety-net
A PreToolUse hook that blocks destructive commands and secret access before AI coding agents run them, parsing command semantics so shell wrappers and flag reordering cannot bypass it.
archestra-ai/OpenAPPA
OpenAPPA is a deterministic policy engine (APPA) placed between an AI agent and its tools that checks every tool call against a declarative data-sensitivity policy before it runs, blocking unauthorized data flows.
wuyoscar/Internal-Safety-Collapse
ISC-Bench is a research project and benchmark for evaluating the "Internal Safety Collapse" vulnerability in LLMs, where models bypass safety controls when completing complex tasks.
cuga-project/cuga-agent
CUGA is an open-source generalist agent harness for enterprises, supporting complex task execution on web and APIs, OpenAPI/MCP integrations, composable architecture, reasoning modes, and policy-aw...
toby-bridges/api-relay-audit
A local security audit tool for AI API relays and LLM proxies, designed to detect prompt injection, model substitution, tool-call rewriting, and other tampering.
hoophq/hoop
Open-source sidecar that sits in front of databases and other resources to mask sensitive data and block destructive queries before AI agents can run them.
decionis/agent-safe-pipeline
Reference architecture and TypeScript library that routes AI agent actions through an independent authorization boundary, turning proposals into policy verdicts, human approvals and single-use grants.
cordum-io/cordum
Source-available control plane that gates AI-agent actions behind approval workflows, a policy safety kernel, and a compliance firewall before they execute.
deadbits/vigil-llm
Vigil is a security scanner for LLM prompts and responses, designed to detect prompt injections, jailbreaks, and other adversarial attacks using various scanning methods like vector databases, YARA...
Justin0504/Aegis
Aegis is a pre-execution firewall for AI agents, providing runtime policy enforcement, cryptographic audit trails, human-in-the-loop approvals, and a kill switch without code changes.
ZaxbyHub/opencode-swarm
OpenCode Swarm is an architect-centric agentic plugin for OpenCode that orchestrates specialized AI agents to generate, review, and test code, enforcing rigorous gated pipelines and security guardr...
SponsioLabs/Sponsio
Sponsio provides deterministic runtime safety solutions for AI agents, enforcing contracts on agent procedures in milliseconds with zero LLM cost.
PrismorSec/prismor
Prismor provides runtime security hooks for AI coding agents, blocking dangerous commands, preventing secret leaks, and defending against prompt injection.
pegasi-ai/reins
Reins provides security controls for AI agents by enforcing deterministic policies, scanning for vulnerabilities, tracking drift with an immutable audit trail, and intervening on risky actions.
agentcontrol/agent-control
Agent Control provides a centralized control plane for enforcing runtime guardrails and safety policies for AI agents, blocking prompt injections, PII leakage, and other risks.
ucsandman/DashClaw
DashClaw is an AI agent governance runtime that intercepts actions, enforces guard policies, manages approvals, and produces audit-ready decision trails for AI agents interacting with real systems.
DobermanCore/Doberman-Core
Local, open-source runtime guardrail that sits between AI coding agents and their tools, giving every action a PASS/AUTH/BLOCK verdict to stop destructive or exfiltrating commands before they run.
praetorian-inc/julius
A Go tool that fingerprints LLM serving infrastructure on network endpoints, identifying which of 60+ AI services — Ollama, vLLM, LiteLLM, TGI and others — is running behind a port.
nizos/probity
Hooks into Claude Code, Codex, and GitHub Copilot CLI to enforce coding rules, such as strict TDD or a ban on destructive commands, by checking every file write and shell command before it runs.
privacera/paig
PAIG (Privacera AI Guardrails) is an open-source framework designed to protect Generative AI applications by ensuring security, safety, and observability for responsible AI deployment.
dodoguardai/dodoguard
DodoGuard is a self-hosted platform (CLI, web console, desktop app) that evaluates LLM prompt quality, red-teams agents for jailbreaks and tool or data abuse, audits agent platforms like Dify and n8n, and produces private release-evidence reports.
openguardrails/openafw
Local proxy that sits in front of Anthropic, OpenAI, and Gemini API calls from coding agents, replacing secrets in outgoing requests with placeholders and restoring them in responses so raw keys never reach the model or provider.