AI guardrails & agent security
3,625 agents · most starred first
AI guardrails & agent security, page 23 of 37
AgentLock authorization middleware for LangChain tool calls.
AgentMesh trust layer integration for LangChain - cryptographic identity and trust-gated tool execution
AgentOath trust protocol integration for LangChain -- automatic trust receipts for chains, tools, and LLM calls.
LangChain integration for AgentOracle — trust verification tools for AI agents
LangChain integration for Agentra — identity oracle and trust layer for autonomous AI agent transactions.
Secure your LangChain agents with per-agent identity, policy enforcement, and tamper-proof audit logs.
Secure LangChain agents with Amorce (Ed25519 + HITL + A2A)
Arc Gate prompt injection detection for LangChain
LangChain Auth SDK for OAuth and tool authentication
Trust enforcement adapter for LangChain and LangGraph — powered by CapiscIO
Cisco AI Defense middleware for LangChain agents — runtime security inspection of LLM inputs/outputs and tool/MCP calls
DeepKeep AI Firewall middleware integration for LangChain
LangChain integration for Σ OVERWATCH — governed, auditable LangChain chains and agents
EDON governance for LangChain agents — govern, audit, and enforce every tool call
LangChain integration for EnigmAgent — resolve {{PLACEHOLDER}} secrets at the LLM boundary so models never see real API keys
EYDII Verify middleware for LangGraph/LangChain — verify every tool call before execution
LangChain integration for ForceField AI security -- scan prompts and moderate outputs in your LangChain pipeline.
Forge Verify + Execute middleware for LangGraph/LangChain — verify tool calls and track executions with cryptographic receipts
Langchain implementation for Guardrails: Nemo
KavachOS DAN gate callback for LangChain agents — pre-execution governance via AEGIS
LangChain integration for Kerq trust scoring and telemetry
LangChain integration for the Kredo agent attestation protocol
KYA (Know Your Agent) identity verification for LangChain agents
LedgerProof adapter for LangChain and LangGraph. Emits EU AI Act Article 50 transparency receipts as a side channel.
Trust-gated MCP tool calls for LangChain via Dominion Observatory
MCPS (MCP Secure) integration for LangChain -- cryptographic identity and trust verification for AI agents.
5-layer security framework for AI agents: identity, capabilities, audit, time-bound permissions, and delegation chains.
LangChain compliance middleware with hash-chained audit trails for EU AI Act readiness
Nuggets authority middleware for LangChain / LangGraph
OpenTerms adapter for LangChain — sign and emit ORS v0.1 receipts for every tool call.
LangChain integrations for Pangea
Access control tools and retrievers for LangChain.
LangChain integration for PlainID authorization
LangChain integration for PQS (Protocol Quality Standard) — Certificate Authority for AI agent economy
LangChain provider for Senthex Proxy — EU-hosted AI firewall with parallel runtime shields.
SidClaw governance integration for LangChain — policy evaluation, human approval, and audit trails for AI agent tools
SpiceDB authorization for LangChain RAG pipelines and agents - fine-grained access control for AI applications
Official SupraWall security integration for LangChain Python
Per-tool input validation and output-boundary policies for LangChain agents.
LangChain tool for behavioral trust scoring via Dominion Observatory
LangChain integration for TruthVouch Trust API — hallucination detection and content verification
LangChain tool that lets agents verify URL safety before navigation via URLCheck MCP server
Vaultak runtime security callback handler for LangChain agents
ZeroTrustedAI library
LangEvals OpenAI moderation evaluator for LLM outputs.
This package is required for integrating your Langchain based Agentic Application to Aryaka's AISecure Guard Service.
LangGraph integration for PlainID authorization
Universal gating and audit layer for AI agents — policy-based tool gating with WebAuthn approval flows
Using Ledger as hardware SSH/GPG agent
Python SDK for Lelu — framework-agnostic authorization engine for AI agents with confidence-aware controls, Docker deployment, and human-in-the-loop workflows
Using OnlyKey as hardware SSH and GPG agent
Using hardware wallets as SSH/GPG/age agent
A python library for requesting OpenID Connect access tokens from oidc-agent.
Dependency-free Python client for a local license-agent, with optional FastAPI enforcement helpers.
Lifeguard integration with OpenAI
Model encryption and authorization extension for vLLM 0.17.0+ on Ascend NPU
Model encryption and authorization extension for vLLM 0.18.0+
Core encryption and license components for vLLM model security
Placeholder package to prevent dependency confusion attacks
HDP (Human Delegation Provenance) — cryptographic authorization provenance for LlamaIndex agents
Vaultak runtime security callback handler for LlamaIndex agents
LlamaIndex integration for ForceField AI security -- scan prompts and moderate outputs in your LlamaIndex pipeline.
llama-index output_parsers guardrails integration
llama-index packs llama_guard_moderator integration
llama-index packs zenguard integration
llama-index packs zenguard guardrails integration
llama-index tools ASIMOV integration
LlamaIndex tools integration for EnigmAgent — resolve {{PLACEHOLDER}} secrets at the LLM boundary so models never see real API keys
EYDII Verify tools for LlamaIndex — verify every agent action before execution
Forge Verify + Execute tools for LlamaIndex — verify agent actions and track executions with cryptographic receipts
TrustGuard integration for LlamaIndex - protect your RAG pipeline from malicious content
LlamaIndex integration for TruthVouch Trust API — hallucination detection and content verification
Remote safety provider for Llama Stack integrating FMS Guardrails Orchestrator and community detectors
LlamaFirewall is a framework designed to detect and mitigate AI centric security risks, supporting multiple layers of inputs and outputs, such as typical LLM chat and more advanced multi-step agentic operations. It consists of a set of scanners for different security risks.
Framework for testing vulnerabilities of GenAI systems.
A package for dehallucinating LLM outputs.
Polymorphic Prompt Assembler to protect LLM agents from prompt injection and prompt leak
Antibodies for LLM hallucinations
LLM Attestation → This package is now `aiir`. Install: pip install aiir
Static security analyzer for LLM applications — eslint for LLM security
Llm Compliance Comparison tools for AI agents. Capabilities: compare providers, recommend for use case, provider risk profile. Built by MEOK AI Labs.
A simple and customizable content filter for LLMs
Black-box LLM fingerprinting system for model identification
LLM-Guard is a comprehensive tool designed to fortify the security of Large Language Models (LLMs). By offering sanitization, detection of harmful language, prevention of data leakage, and resistance against prompt injection attacks, LLM-Guard ensures that your interactions with LLMs remain safe and secure.
v0.66.0: sycophancy_score() direction corrected — AUROC 0.9296 [0.8754, 0.9703] on Perez 2022 (was 0.0704, wrong direction). AgentShield judge_risk field fix.
LLM Guard — PII 탐지 및 프롬프트 인젝션 차단 라이브러리
AI safety guardrail — intent analysis, prompt injection detection, and policy enforcement for LLM applications
A package for LLM guardrails
Drop-in prompt injection defense for LLM apps and AI agents — detect, sanitize, block, and audit injection attacks in real time. Includes multi-turn session scanning, allow-lists, rate-abuse detection, multi-layer scanner, FastAPI and Flask middleware.
Deterministic Unicode-aware input hardening for LLM applications.
A package for sanitizing and securing user inputs to LLMs
LLM Integrity → This package is now `aiir`. Install: pip install aiir
Guardrails for LLM-generated medical and health content
A tool for obfuscating text by manipulating token IDs while preserving token count and structure
Validate LLM outputs against schemas with automatic retry and JSON extraction.
Sanitize LLM outputs before HTML, SQL, shell, or markdown sinks. Python port of @mukundakatta/llm-output-sanitizer.
Security testing toolkit for LLM-based systems
Lightweight prompt injection detection and blocking
Reliable, structured, production-safe LLM outputs with schema validation and auto-repair
CLI + MCP server for detecting, classifying, and redacting embedded LLM agent instructions in documents, source code, and web pages.