Prompt Injection Defense Python Packages

vmware-policy

Unified audit logging, policy enforcement, and sanitization for VMware MCP skills

guardlayer

Runtime guardrails between an AI agent and its tools: scan what it reads, gate what it does, and track the session so a secret read earlier can't be exfiltrated later. Containment over detection — pure Python, zero dependencies, ~0.2 ms per tool call.

agentlock

An adversarially benchmarked reference implementation for pre-action AI agent authorization. Provenance-based gating for LLM agent tool calls: deny-by-default permissions, parameter lineage, signed receipts, audit logging.

grey-panda

The calm guardian for AI, agent, and MCP code — a zero-dependency, OWASP-anchored AI security toolkit.

capability-reasoning-kernel

Python reference for a CaMeL-like Reasoning Kernel. Control/data-flow separation reduces prompt-injection-to-effect risk when planner input is trusted, the capability policy is correct, and tools enforce it. Topology, not a proof that injection cannot cause an unauthorized effect.

patronus-api-client

Security scanner and runtime protection for AI coding agents. Detects prompt injection, secrets, PII, and unsafe tool activity across Codex, Claude Code, and DeepSeek Harness.

verb-authority

Prevent untrusted data from authoring protected tool-call arguments.

decoyshield

Counter-recon honeypot against agentic LLM attackers (PentestGPT/AutoGPT/LangChain). Six deploy modes: Flask drop-in, WSGI/ASGI middleware (Django/FastAPI), callable Python primitives, decoyshield CLI, edge configs (nginx/Caddy/Cloudflare), and Node.js npm package (Express/Fastify). Invisible-to-human, visible-to-LLM bait halts AI scans.

occludra

Self-hosted AI security proxy. Redact PII, block prompt injection, route to any LLM provider. OpenAI-compatible.

langguard

LangGuard Python Library

aisg

Self-hosted AI security proxy. Redact PII, block prompt injection, route to any LLM provider. OpenAI-compatible.

sec-guardrails

Defensive 7-layer runtime guardrails gateway for LLM agents: a reverse proxy that blocks prompt injection (incl. indirect/XPIA), tool misuse, and data exfiltration on every turn. Polyglot — Python control plane + Rust security core + React HITL. Metrics reported split (ASR vs FPR), never one blended F1.

pulsarai

An AI security library for block prompt injection and more 🤖👾

zeos

ZEOS: a transformer operating system — priority-preemptive scheduling, protected mode, and deterministic replay for LLM jobs

vajraclaw

⚡ DROS™ VajraClaw (Hacker Edition): Free Multi-Agent Runtime Security & Docker Governance Gateway with W3C DID & <1μs Enforcement

promptscreen

Protect your LLMs from prompt injection and jailbreak attacks. Easy-to-use Python package with multiple detection methods, CLI tool, and FastAPI integration.

agent-trap

Counter-recon honeypot against agentic LLM attackers (PentestGPT/AutoGPT/LangChain). Six deploy modes: Flask drop-in, WSGI/ASGI middleware (Django/FastAPI), callable Python primitives, decoyshield CLI, edge configs (nginx/Caddy/Cloudflare), and Node.js npm package (Express/Fastify). Invisible-to-human, visible-to-LLM bait halts AI scans.