Add OWASP Agent Memory Guard - AI agent memory security (OWASP ASI06)

[vgudur-dev]

Summary

Adding OWASP Agent Memory Guard — the reference implementation for OWASP ASI06 (Memory Poisoning) in the OWASP Top 10 for Agentic Applications.

What it does

• Detects and blocks memory poisoning attacks in AI agents at runtime
• Scans memory reads/writes for prompt injection payloads, secret leakage, and structural tampering
• Sub-100μs latency, zero external dependencies
• Adopted by the UK Government BEIS inspect_evals framework

Install

pip install agent-memory-guard
# or for LangChain
pip install langchain-agent-memory-guard

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

[JinyangWang27]

@vgudur-dev Thanks for your submission. However, this project does not meet our acceptance criteria:

GitHub stars: 4 stars (minimum 100 required, or strong justification for Hidden Gem).

Please see our CONTRIBUTING.md for full requirements. You're welcome to resubmit once the project has matured and gained community traction.