langchain - 💡(How to fix) Fix [Feature Request] Built-in Memory Poisoning Protection via OWASP Agent Memory Guard

Problem

AI agents using LangChain's memory are vulnerable to memory poisoning attacks — where adversarial inputs injected into conversation history or RAG stores cause the agent to leak secrets, ignore instructions, or behave maliciously. OWASP has identified this as a top risk for LLM applications.

Proposed Solution

OWASP Agent Memory Guard (AMG) is an open-source Python library that wraps any memory store as a transparent security layer:

• pip install agent-memory-guard
• Scans every memory write for prompt injection, PII leakage, and tampering
• 92.5% detection rate on AgentThreatBench
• Zero-config: works with LangChain's BaseChatMessageHistory

Example

from langchain_community.chat_message_histories import ChatMessageHistory
from agent_memory_guard import MemoryGuard

history = ChatMessageHistory()
guard = MemoryGuard(history)
guard.add_message(HumanMessage(content="..."))

Links

• GitHub: https://github.com/OWASP/www-project-agent-memory-guard
• PyPI: https://pypi.org/project/agent-memory-guard/
• Benchmark: https://pypi.org/project/agent-threat-bench/

Would the LangChain team consider adding AMG as recommended security middleware in the docs? Happy to contribute a PR.