OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Security & Sandboxing

Agentic Radar

Agentic Radar is a Python command-line security scanner for agentic workflows. It analyzes workflow code, visualizes agent structure, identifies tools and MCP servers, and creates HTML reports with vulnerability mappings. Its separate vulnerability-testing and prompt-hardening features support selected agent frameworks.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Agentic Radar — OpenViST project profile in Agent Security & Sandboxing.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Agentic Radar when you need a code-oriented view of an agent workflow’s structure and potential security issues, or when adding vulnerability testing for an OpenAI Agents workflow.

Who it is for

  • Developers reviewing agent workflows before deployment
  • Security professionals assessing tools and possible vulnerabilities in agent systems
  • Teams that want workflow security checks in development or CI/CD processes

Documented uses

  • Scan workflow code and create an HTML security report.
  • See workflow structure and identify external or custom tools and MCP servers.
  • Review mappings between detected tools and known vulnerabilities.
  • Test OpenAI Agents workflows for vulnerabilities.
  • Improve detected system prompts and include the hardened prompts in a report.
Before you choose

Requirements & limits

The project is open source under the Apache-2.0 license and is installed as a Python package with pip. Vulnerability testing requires an OPENAI_API_KEY environment variable.

  • The CLI scan command documents LangGraph, CrewAI, n8n, OpenAI Agents and AutoGen as framework identifiers.
  • The vulnerability-testing command is documented for OpenAI Agents and requires an OPENAI_API_KEY.
  • CrewAI extras are needed for its vulnerability-testing feature and predefined tool descriptions; crewai-tools supports Python versions >=3.10 and
  • CrewAI tool descriptions may be less detailed or missing when using a Python version outside that range.
  • The README lists prompt hardening for OpenAI Agents, CrewAI and AutoGen.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install Agentic Radar

Install the CLI and check that it is available.

Before you start

  • Python with pip

Steps

  1. Install the package with the command below.

    Documented command
    pip install agentic-radar

  2. Run agentic-radar --version to check the installation.

Expected outcome

The Agentic Radar command-line tool is installed and its version check can be run.

Guide limitations

  • Some framework-specific features require additional installation extras.
Documentation 1 ↗
Source-based practical guide

Scan a LangGraph project and save a report

Generate an HTML report from a LangGraph workflow directory.

Before you start

  • Python with pip and Agentic Radar installed
  • A LangGraph project directory to scan

Steps

  1. Run agentic-radar scan langgraph -i path/to/langgraph/example/folder -o report.html.

  2. Review the generated HTML report.

Expected outcome

An HTML report file is requested at `report.html` for the specified workflow directory.

Guide limitations

  • The documented example demonstrates LangGraph scanning; other frameworks have their own framework identifiers.
  • The guide describes the documented command, not a verified execution.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
12 October 2026
Original website title
splx-ai/agentic-radar · Official homepage title · retained in its source language
How we review sourcesSuggest a correction

Related options

Agentic Visualizer — A separate web-based tool identified in the Agentic Radar README for users who only need to visualize agent workflows.