Neura Market
agentic-radar logo

agentic-radar

Free

A security scanner for your LLM agentic workflows

CybersecurityFreeFree tier
Inputs: text, codeOutputs: text, file
Type
Open Source
Company
SPLX
Visit Website GitHub (960 stars)

About agentic-radar

Agentic Radar is an open-source security scanner developed by SPLX (now part of Zscaler) specifically designed for agentic AI workflows. It is described as the first static application security testing (SAST) tool for agentic systems, enabling users to visualize agent workflows, identify attack patterns and tool-level vulnerabilities, and receive hardening recommendations for each agent. The tool is part of the broader SPLX platform, which provides end-to-end security for AI systems, including red teaming, runtime protection, governance, and remediation. Agentic Radar is available for free on GitHub, targeting developers and security teams who need to assess and secure their LLM-based agentic applications.

Key Features

Visualizes agentic AI workflows to map out tool interactions and data flow
Surfaces attack patterns and tool-level vulnerabilities specific to agentic systems
Provides hardening recommendations for each identified agent
Open-source and free to use, with source code available on GitHub
Designed as a SAST (static application security testing) tool for agentic workflows
Part of the SPLX platform, which offers broader AI security capabilities including red teaming, runtime protection, and governance

Pros & Cons

Pros
  • Open-source and free, with no apparent licensing costs
  • Specifically tailored for agentic AI workflows, addressing a niche security need
  • Provides actionable hardening recommendations for each agent
  • Backed by SPLX (now part of Zscaler), indicating enterprise-grade security expertise
  • Visualizes workflows, aiding in understanding complex agent interactions
Cons
  • Limited to static analysis; runtime vulnerabilities may not be detected
  • Effectiveness depends on the accuracy of workflow visualization and vulnerability detection algorithms
  • Requires technical expertise to interpret results and implement hardening recommendations
  • As an open-source tool, support and updates may depend on community contributions and SPLX's ongoing maintenance

Best For

Security auditing of LLM-based agentic workflows during developmentIdentifying vulnerabilities in multi-agent systems and tool integrationsHardening agentic AI applications against prompt injections and other attacksIntegrating security scanning into CI/CD pipelines for agentic AI projectsEducational use for understanding attack surfaces in agentic architectures

Alternatives to agentic-radar

giskard-oss

🐢 Open-Source Evaluation & Testing library for LLM Agents

h4cker

This repository is maintained by Omar Santos (@santosomar) and includes thousands of resources related to ethical hacking, bug bounties, digital forensics and incident response (DFIR), AI security, vu

ai-captcha-bypass

AI Captcha Bypass

backdoor-learning-resources

A list of backdoor learning resources

claude-bug-bounty

AI-powered bug bounty hunting from your terminal - recon, 20 vuln classes, autonomous hunting, and report generation. All inside Claude Code.

FAQ

What is Agentic Radar?
Agentic Radar is an open-source security scanner for agentic AI workflows, developed by SPLX. It visualizes agent workflows, identifies attack patterns and vulnerabilities, and provides hardening recommendations.
Is Agentic Radar free to use?
Yes, based on available information, Agentic Radar is open-source and free. The source code is available on GitHub, and the SPLX website describes it as free.
How does Agentic Radar differ from other security scanners?
It is specifically designed for agentic AI workflows, focusing on static analysis of agent interactions, tool-level vulnerabilities, and attack patterns, unlike general-purpose scanners.
Does Agentic Radar require integration with the SPLX platform?
Based on available information, Agentic Radar is a standalone open-source tool, but it is part of the broader SPLX platform. Integration details should be verified on the GitHub repository.
What types of vulnerabilities can Agentic Radar detect?
It surfaces attack patterns and tool-level vulnerabilities specific to agentic systems, such as insecure tool configurations and potential prompt injection vectors. Exact detection capabilities should be checked in the documentation.
Who is the target audience for Agentic Radar?
Developers, security engineers, and AI practitioners who build or deploy LLM-based agentic applications and need to assess their security posture.