Security & AI safety

AI guardrails & agent security

3,625 agents · most starred first

AI guardrails & agent security, page 31 of 37

navige

AI governance SDK — intercept, audit, and control AI agent tool calls. Add human-in-the-loop approval, kill switches, and a full audit trail to any AI agent.

@idevsec/creduent-cli

The official command-line interface for Creduent: the open standard and security protocol for cryptographic AI agent identity, authentication, and trust verification using Ed25519 and DNS.

@pushary/langgraph

Phone approvals for LangGraph and LangChain agents. Your agent asks, your user taps Approve or Deny.

@tryagent/adapters

Framework adapters that wire TryAgent human-in-the-loop guardrails into LangGraph, the Vercel AI SDK, and the OpenAI Agents SDK.

toolgovern-cli

CLI for toolgovern -- validate policy files, audit local gate traces, and scaffold framework integration boilerplate, with a --json output mode for programmatic/agent use.

0din-litellm-shield

SusFactor jailbreak/prompt-injection guardrail for LiteLLM

AIAgentFirewall

A security firewall for AI agents, skills and MCP servers — scan before you install.

aegis-agentguard

AgentGuard installation entry point

aegis-agentguard-api

Guard API / Control Plane service for AgentGuard

aegis-agentguard-cli

Headless command-line control utility for AgentGuard

aegis-agentguard-core

Stateless AgentGuard Core security decision engine

aeonic-agentguard-sdk-python

Official Python SDK for Aeonic — runtime monitoring and governance for AI agents

agent-authz

A secure-by-default authorization PEP for APIs, Agent tools, RAG, and tasks

agent-callguard

Guardrails for AI-agent tool/function calls: schema validation, realistic error recovery, cost/quota budgets, circuit breakers, and response caching. Framework-agnostic.

agent-circuit-breaker

Deterministic safety gate for AI coding agents

agent-control-evaluator-defenseclaw

DefenseClaw evaluators for agent-control

agent-control-spec

Agent Control Specification: a stateless policy decision runtime implementing the agent-hooks interceptor contract.

agent-core-credentials

KeePass-backed credential vault for agent-core agents

agent-core-inbound

Inbound notifications router — deny-by-default external-event surface for agent-core beings.

agent-diff-guard

Flags risky changes to AI agent configs (prompts, tool permissions, model choice) in a diff.

agent-gorgon

User-space runtime policy guard with deterministic control decisions and forensic evidence.

agent-governance-toolkit-compliance

Public Preview — Unified installer and runtime policy enforcement for the Agent Governance Toolkit

agent-governor

Embedded multi-tenant Agent governance engine - cost attribution, declarative policies, and audit

agent-hooks-sdk

Framework-neutral agent control contract: interception points, agent context, verdict, and conformance test kit. Python wrapper over the canonical Rust core. Cooperative contract, not a security boundary: the host is fully trusted (see SECURITY.md).

agent-input-sanitizer

Python client for agent-input-sanitizer: a thin bridge to the Node.js CLI for byte-identical sanitization verdicts.

agent-loop-guard-runtime

Local runtime guard for coding agents with loop detection and explainable policy decisions.

agent-passport-agl

Python SDK for Agent Passport - AI Agent compliance proof system for EU AI Act Art.50

agent-runtime-governance

Action Commit Safety for AI agents: exact-call approval, UNKNOWN outcome recording, idempotent execution, and signed audit.

agent-sandbox-backends

OpenSandbox backend SDK for AI agents

agent-toolbox.ai

Python SDK for agent-toolbox.ai — deterministic pre-action gates for AI agents

agent-tripwire

Safety checks an AI agent runs before it acts: is this package malware/typosquat, is this shell command destructive, does this text leak a secret. CLI + MCP server.

agent-zero-trust

Offline repository inspection for coding-agent instructions: scan, compare, explain, and export a local review.

agentcikit

CI-grade evidence and safety tools for AI agents, MCP servers, and open-source contribution workflows.

agentctl-sdk

Official Python SDK for agentctl — sign, hash-chain, and govern every agent action. Zero dependencies.

agentfences

Runtime governance for AI agents — budget limits, loop protection, token limits, audit trail

agentguard-dev

An honest pre-deployment reliability and security gate for AI agents.

agentic-guard-mcp

Static security analysis for MCP servers and AI-agent workspace configurations. Taint-graph based detection of tool-description injection, capability overreach, and exfiltration paths.

agenticstack-guard-secrets

Secrets detection and redaction guard plugin for AgenticStack

agentmandate

Compound-path and cross-release analysis of what an AI agent is permitted to do

agentpolicypack

A vendor-neutral policy-as-code toolkit for governing AI agents and agentic workflows.

agentproof-scan

CLI security scanner for AI agents: probes your HTTP endpoint (or scans traces you already have) for leaked secrets and system prompts, including the reasoning channel. Runs keyless with --demo; CI-gating exit codes.

agentrust-capture-core

Shared fingerprinting, comparison and baseline-sealing core for AgenTrust agent-integrity capture engines

agentsecure-sdk

AgentSecure SDK — LLM DLP proxy for Python

agenttic

Agenttic — certify, trace and enforce AI agents. One install, one import, a signed safety grade.

agentvigilante

Local containment runtime and firewall for AI coding agents

argorix-guardrails-agent

Official Argorix Agent Guardrails SDK for Python agents and tool execution

bounded-agent

Contract-bounded autonomy for LLM agents: five small primitives that make an agent safe to leave alone.

clavenar-agent-sdk

Wrap your Anthropic / OpenAI Python client. Every tool call is inspected by Clavenar before it runs.

comply-agent

AI Agent OWASP Agentic Top 10 Compliance Scanner

custodian-claude-guard

Fail-closed Custodian policy guard for Claude Code, with unstrippable managed-hook enforcement

cyber-agents

AI-security framework — build agents that are safe by construction (placeholder, under active development)

ellm-guardrail

Python SDK for the ELLM Guardrail — deterministic policy firewall for AI agent tool calls

fairllms

Fairness definitions and bias metrics for large language models

fidacy-langchain

Gate LangChain tool calls behind a Fidacy verdict. Wraps the tool itself, so a denied payment never executes: callbacks can only observe, this can stop.

foresight-agent-guard

Predictive cost, policy, and runaway-session guardrails for AI coding agents

guardrails-ai-toxic-language-llm

Detects toxic language in LLM-generated text using an LLM as the detection backbone. Evaluates text across seven toxicity categories: toxicity, severe toxicity, obscene, threat, insult, identity attack, and sexual explicit content.

hallmark-identity

Open, IDP-agnostic agent-identity toolkit: agent + on-behalf-of tokens via RFC 8693.

hemlock-rag

AI security lab — attack and defense testing for RAG pipelines and tool-using agents

inter-agent-guard

Inter-agent security firewall for multi-agent AI systems

kratos-agentrails

A domain-agnostic safety layer for AI agents: declarative policy, dry-run, audit ledger and circuit breaker between an agent's proposed actions and their execution.

langchain-aidress

Aidress agent trust and discovery tools for LangChain

langchain-axiorank

AxioRank middleware for LangChain agents: govern every tool call and model turn through one Zero-Trust security gateway.

langchain-fidacy

Fidacy action firewall for LangChain tool calls. Convention-named entry point: the implementation lives in fidacy-langchain, and this package re-exports it unchanged.

langchain-preloop

LangChain integration for Preloop — the open-source AI agent control plane (MCP firewall, model gateway, human approvals, budgets, and a tamper-evident audit ledger).

langchain-relayshield

LangChain tools and a mandatory pre-execution gate for RelayShield's MCP registry risk and prompt-injection breach checks.

llm-guardrails-sk

Provider-agnostic middleware for LLM safety

llm-pg

Runtime prompt-injection defense middleware for LLM applications

llm-scrub

Zero-dependency scrubbers for LLM chat output — strip leaked JSON state-op fences, backend scaffold blocks, and tool-preamble meta-narration from player-facing text

llm-security-pipeline

Defensive pipeline for LLM apps: prompt injection, tool-call gating, output/PII guards

llmfw

LLM Firewall: a safety layer between your application and any LLM API — input/output firewalling, PII redaction, jailbreak detection, and domain-specific guardrails.

llmsafespaces

Python SDK for LLMSafeSpaces API

llmuid

Identifiers designed to survive being read, copied and re-emitted by language models

maigp-anthropic

AIGP governance adapter for anthropic

maigp-autogen

AIGP governance adapter for Microsoft AutoGen multi-agent conversations

maigp-crewai

AIGP governance adapter for CrewAI multi-agent orchestration

maigp-llamaindex

AIGP governance adapter for llamaindex

maigp-openai

AIGP governance adapter for openai

mcp-bastion-autogen

MCP-Bastion security helpers for Microsoft AutoGen / Agent Framework tool messages.

mcp-bastion-langgraph

MCP-Bastion security helpers for LangGraph nodes and tool calls.

mcp-bastion-litellm

MCP-Bastion security middleware for LiteLLM (100+ LLM providers via one OpenAI-compatible API).

mcp-bastion-ollama

MCP-Bastion security middleware for Ollama local models (OpenAI-compatible API).

mcp-bastion-openai-agents

MCP-Bastion security helpers for OpenAI Agents SDK tools and messages.

openagent-control

Agent identity & governance control plane: policy-gated, credential-brokered, cryptographically audited tool execution for autonomous agents.

openagent-oas

Python SDK for the Open Agent Specification (OAS) — decentralized identity for autonomous entities

openagentid-sdk

OpenAgent SDK — identity, authentication, and Agent Capability Token verification for autonomous agents.

openai-agents-relayshield

OpenAI Agents SDK tools and a mandatory pre-execution gate for RelayShield's MCP registry risk and prompt-injection breach checks.

openguardrails-instrumentation-langgraph

Guard a custom LangGraph agent with the OpenGuardrails (OGR) protocol — a drop-in ToolNode at the agent-hook altitude, with native LangGraph interrupt() as the human-confirm gate.

ory-agent-framework

Ory Agent Security for the Microsoft Agent Framework — per-tool authorization, activity logging, and identity propagation via function middleware. Built on ory-argus.

ory-crewai

Ory Agent Security for CrewAI — per-tool authorization, activity logging, and identity propagation by wrapping tools. Built on ory-argus.

ory-langchain

Ory Agent Security for LangChain / LangGraph — per-tool authorization, activity logging, and identity propagation via an AgentMiddleware. Built on ory-argus.

ory-llamaindex

Ory Agent Security for LlamaIndex — per-tool authorization, activity logging, and identity propagation by wrapping FunctionTools. Built on ory-argus.

ory-openai-agents

Ory Agent Security for the OpenAI Agents SDK — per-tool authorization, activity logging, and identity propagation via a tool-input guardrail. Built on ory-argus.

ory-pydantic-ai

Ory Agent Security for Pydantic AI — per-tool authorization, activity logging, and identity propagation via a WrapperToolset. Built on ory-argus.

qarai-agent-guard-crewai

CrewAI integration for qarai-agent-guard that provides runtime security for AI agents through threat detection, policy enforcement, and content protection.

qarai-agent-guard-langchain

LangChain integration for qarai-agent-guard that provides runtime security for AI agents through threat detection, policy enforcement, and content protection.

rag-inject-guard

RAG indirect prompt injection guard: scan retrieved documents for prompt-injection (instruction override, system/tool-prompt manipulation, exfiltration, invisible-Unicode/homoglyph smuggling) in English and Turkish, and quarantine poisoned docs before they reach the model.

securellm-agentguard

🛡️ Experimental security guardrails for LangChain agent code execution — Docker sandbox isolation.

sentinel-agent-trust

Sentinel AI — client SDK for AI agent governance & identity. Sign agent actions with Ed25519 and enforce them through the Sentinel authorization gateway. Private beta: [email protected]

sentinel-agent-trust-crewai

CrewAI adapter for Sentinel AI agent governance — wrap any CrewAI tool so every call is authorized (and audited) by the Sentinel gateway before it runs. Private beta: [email protected]

sentinel-agent-trust-langchain

LangChain adapter for Sentinel AI agent governance — wrap any LangChain tool so every call is authorized (and audited) by the Sentinel gateway before it runs. Private beta: [email protected]

More in Security & AI safety