8000 GitHub - nicho2/agentic-radar: A security scanner for your LLM agentic workflows
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

A security scanner for your LLM agentic workflows

License

Notifications You must be signed in to change notification settings

nicho2/agentic-radar

 
 

Repository files navigation

logo

A Security Scanner for your agentic workflows!

contributors last update forks stars open issues license

Table of Contents
  1. Description
  2. Getting Started
  3. Usage
  4. Roadmap
  5. Demo
  6. Contributing
  7. Code Of Conduct
  8. License

Description 📝

The Agentic Radar is designed to analyze and assess agentic systems for security and operational insights. It helps developers, researchers, and security professionals understand how agentic systems function and identify potential vulnerabilities.

It allows users to create a security report for agentic systems, including:

  1. Workflow Visualization - a graph of the agentic system's workflow✅
  2. Tool Identification - a list of all external and custom tools utilized by the system✅
  3. Vulnerability Mapping - a table connecting identified tools to known vulnerabilities, providing a security overview✅

The comprehensive HTML report summarizes all findings and allows for easy reviewing and sharing.

Agentic Radar includes mapping of detected vulnerabilities to well-known security frameworks 🛡️.

Why Use It? 🔎

Agentic systems have complex workflows and often interact with multiple tools, making transparency and security assessment challenging. This tool simplifies the process by offering a structured view of workflows, tools, and potential risks.

Detailed Report

Getting Started 🚀

Prerequisites

Agentic Radar uses graphviz and cairo for dependency graph visualization.

  • graphviz
# homebrew
brew install graphviz

# linux
apt-get install graphviz
  • cairo
# homebrew
brew install cairo

# linux
apt-get install libcairo

Installation

pip install agentic-radar

# Check that it is installed
agentic-radar --version

Usage

Run agentic-radar --help for more info:

Usage: agentic-radar [OPTIONS] COMMAND [ARGS]...

Options:
  -i, --input-dir TEXT            Path to the directory where all the code is
                                  [env var: AGENTIC_RADAR_INPUT_DIRECTORY;
                                  default: .]
  -o, --output-file TEXT          Where should the output report be stored
                                  [env var: AGENTIC_RADAR_OUTPUT_FILE;
                                  default: report_20250311_122338.html]
  --version
  --install-completion [bash|zsh|fish|powershell|pwsh]
                                  Install completion for the specified shell.
  --show-completion [bash|zsh|fish|powershell|pwsh]
                                  Show completion for the specified shell, to
                                  copy it or customize the installation.
  --help                          Show this message and exit.

Commands:
  langgraph  Scan code written with LangGraph
  crewai     Scan code written with CrewAI

Roadmap 📈

Planned features (in no particular order)

Demo 🎥


Demo Google Colab Notebook 📘

Designed for AI engineers and security researchers, this demo showcases how to integrate Agentic Radar into your development workflow. ⚙️

It helps you understand agentic system behavior, visualize security risks, and enhance AI transparency in your applications. 🚀

Community 🤝

We welcome contributions from the AI and security community! Join our Discord community to connect with other developers, discuss features, get support and contribute to Agentic Radar 🚀

If you like what you see, give us a star! It keeps us inspired to improve and innovate and helps others discover the project 🌟

Frequently Asked Questions ❓

Q: Is my source code being shared or is everything running locally?
A: Everything is running locally. Your source code is never uploaded anywhere.

Contributing 💻

CONTRIBUTING

Code Of Conduct 📜

CODE OF CONDUCT

License ⚖️

LICENSE

About

A security scanner for your LLM agentic workflows

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 77.8%
  • Jinja 22.2%
0