
Agentguardx
An open-source red-teaming toolkit for testing the security of LLM-powered agents, auditing prompt injection, memory poisoning, and tool abuse.
About
AgentGuardX provides an open-source AI security toolkit for testing LLM-powered agents. It acts as a red team partner, identifying vulnerabilities like prompt injection, memory poisoning, and tool abuse before they can be exploited by attackers. This toolkit is built on the BSC Chain and serves developers and organizations building and deploying AI agents.
Key facts
At-a-glance profile data
- Website
- agentguardx.xyz
- Founded
- 2026
- Total funding
- -
- Business model
- B2B
Tech stack
Detected on agentguardx.xyz, last scanned 24 Sept 2026. Run your own scan
- Compliance
- SOC 2, ISO 27001
- Frontend
- Bootstrap
- CDN
- Vercel
Add first review
Frequently asked
Is Agentguardx trustworthy and reputable?
StartupHub's Data Trust & Reputation score for Agentguardx is 34 out of 100, based on site security posture and privacy practices.