Nobody owns this listing yet

OWASP Agent Memory Guard is live in the directory, but no verified owner controls the page.

Claim this listing
  • Pick your launch daySchedule a launch and compete for the daily #1 spot.
  • Win trophies you can embedRank top 3 on launch day and get an embeddable trophy.
  • “Featured on AI Kaptan” badgeOwner-only badge with your live upvote count.
  • Control what the page saysEdit copy, pricing, links and screenshots as the owner.

Unclaimed listings don't stay up forever. Listings left unclaimed are removed from the directory in periodic cleanups. When a listing goes, it goes everywhere — the tool page and every alternatives comparison it appears on, along with the rankings and backlinks that page had built up.

Verify with a work email on the domain (about 2 minutes), then a one-time $5 listing fee. See everything you get

Scheduled launch — Aug 19, 2026

OWASP Agent Memory Guard is listed on AI Kaptan and will enter the daily launch competition on that day. Community upvotes open on launch day.

OWASP Agent Memory Guard
AI Security
Free

OWASP Agent Memory Guard

OWASP Agent Memory Guard is an open-source security tool protecting AI agents from memory poisoning attacks by screening memory store operations with detector pipelines and declarative policies.

Rating
0
Reviews
0
Upvotes
about 4 hours ago
Listed
AI security
OWASP
Open Source
Agent Security
Memory Poisoning Protection
Runtime Defense
LLM Guardrails
Tool information
Provider
OWASP Foundation
Platforms
GitHub
Linux
macOS
Windows
Languages
English
API Available
Yes
Added to directory
about 4 hours ago
Last updated
about 4 hours ago
Rating
Not available
Pricing
Free

About OWASP Agent Memory Guard

What the tool does and who it's for

OWASP Agent Memory Guard (www-project-agent-memory-guard) is an open-source software security tool created to protect AI agents from memory-based threats and memory poisoning attacks. Since AI agents persist memory across sessions, malicious or poisoned inputs committed to memory become privileged instructions on subsequent operational turns. Agent Memory Guard sits between the AI agent and its memory store, screening every memory operation through a specialized detector pipeline and declarative policy. This OWASP Foundation project specifically addresses ASI06 (Memory & Context Poisoning), preventing AI agents from being weaponized, hijacked, or exploited for data exfiltration even across system context resets.

Key capabilities

Runtime defense against AI agent memory poisoning attacks
Interception pipeline positioned between the AI agent and its memory store
Detector pipeline for real-time screening of all memory read and write operations
Declarative policy engine for flexible memory security enforcement
Persistent security that protects across system context resets
Direct defense targeting OWASP Top 10 for LLM/Agentic AI threat ASI06 (Memory & Context Poisoning)
Prevents compromised instructions, data exfiltration, and hijacked tool calls

Pricing

Free — plans below

Open Source

Most popular

$0

  • Free and open-source under OWASP Foundation
  • Full source code access
  • Detector pipeline and declarative policy enforcement
  • Community support and updates

Reviews

Be the first to review OWASP Agent Memory Guard

No reviews yet

Be the first to share your experience with OWASP Agent Memory Guard.

Featured Tools

Handpicked by our team of experts

Suggest an Update

Found outdated information? Help us keep this listing accurate.

FAQ

OWASP Agent Memory Guard FAQs

Common questions about OWASP Agent Memory Guard