Agentic Radar
A security scanner for your LLM agentic workflows
What is Agentic Radar?
Agentic Radar is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to security scanner for your llm agentic workflows
A security scanner for your LLM agentic workflows
This server falls under the Security category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- A security scanner for your LLM agentic workflows
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx agentic-radarConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Agentic Radar
Agentic Radar is an open-source security scanner purpose-built for LLM agentic workflows, analyzing code written with frameworks like OpenAI Agents, CrewAI, LangGraph, Autogen, and n8n to identify prompt injection risks, PII leakage, excessive permissions, and other vulnerabilities specific to AI agent architectures. It generates comprehensive HTML security reports with workflow visualizations, maps findings to OWASP and AI security frameworks, and can even perform runtime testing to detect live vulnerabilities in deployed agents. Security engineers and DevSecOps teams use it to bring structured security review to AI systems before they reach production.
Prerequisites
- Python 3.9+ with pip
- An agentic workflow codebase using a supported framework: OpenAI Agents, CrewAI, LangGraph, Autogen, or n8n
- OPENAI_API_KEY (required for prompt hardening and runtime vulnerability testing features)
- Optional: framework-specific extras for full feature support (e.g. pip install 'agentic-radar[crewai]')
Install Agentic Radar
Install the base package via pip. For full support of specific frameworks, install the framework-specific extras.
pip install agentic-radar
# Framework-specific extras (optional but recommended):
pip install "agentic-radar[crewai]"
pip install "agentic-radar[openai-agents]"Set your OpenAI API key
Export your OpenAI API key for features that require LLM access, such as prompt hardening and vulnerability testing.
export OPENAI_API_KEY=your_openai_api_key_hereScan your agentic workflow for vulnerabilities
Run the scan command against your agent codebase directory. Specify the framework type and an output HTML report path.
agentic-radar scan langgraph -i ./path/to/your/workflow -o security-report.html
# For CrewAI:
agentic-radar scan crewai -i ./my-crewai-project -o report.html
# For OpenAI Agents:
agentic-radar scan openai-agents -i ./agents/ -o report.htmlRun runtime vulnerability tests
Use the test command to perform live vulnerability assessment against a deployed agent, checking for prompt injection, PII leakage, and harmful content generation.
agentic-radar test openai-agents "path/to/agent_example.py"Enable prompt hardening in scan reports
Add the --harden-prompts flag to get AI-generated suggestions for making your agent's system prompts more resistant to injection attacks.
agentic-radar scan openai-agents --harden-prompts -i examples/ -o hardened-report.htmlReview the HTML security report
Open the generated HTML report in a browser to review the workflow visualization, identified vulnerabilities, affected tools, external service connections, and OWASP framework mappings.
open security-report.html # macOS
# or: xdg-open security-report.html # LinuxAgentic Radar Examples
Client configuration
MCP server configuration for Agentic Radar when integrated as an MCP server tool in Claude Desktop.
{
"mcpServers": {
"agentic-radar": {
"command": "npx",
"args": ["agentic-radar"],
"env": {
"OPENAI_API_KEY": "your_openai_api_key_here"
}
}
}
}Prompts to try
Example CLI commands and AI assistant prompts for using Agentic Radar to audit your AI workflows.
- Run: agentic-radar scan langgraph -i ./my-agent -o report.html
- Run: agentic-radar test openai-agents ./examples/email_agent.py
- "Scan my CrewAI workflow in ./src for prompt injection vulnerabilities"
- "Generate a hardened version of the system prompts in my agent codebase"
- "Show me all external services my LangGraph agent is connecting to"Troubleshooting Agentic Radar
Scan produces no findings or empty report
Ensure you are specifying the correct framework type (langgraph, crewai, openai-agents, autogen, n8n) and that the -i path points to a directory containing actual agent source code, not an empty folder.
Prompt hardening fails with authentication error
The --harden-prompts flag requires a valid OPENAI_API_KEY. Verify the key is exported correctly with 'echo $OPENAI_API_KEY' and that it has sufficient credits.
Framework-specific features not working (e.g. CrewAI tool detection missing)
Install the framework-specific extra: 'pip install agentic-radar[crewai]' or 'pip install agentic-radar[openai-agents]'. The base installation may not include all framework-specific parsers.
Frequently Asked Questions about Agentic Radar
What is Agentic Radar?
Agentic Radar is a Model Context Protocol (MCP) server that security scanner for your llm agentic workflows It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Agentic Radar?
Follow the installation instructions on the Agentic Radar GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Agentic Radar?
Agentic Radar works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Agentic Radar free to use?
Yes, Agentic Radar is open source and available under the Apache-2.0 license. You can use it freely in both personal and commercial projects.
Agentic Radar Alternatives — Similar Security Servers
Looking for alternatives to Agentic Radar? Here are other popular security servers you can use with Claude, Cursor, and VS Code.
Casdoor
★ 13.6kAn open-source Agent-first Identity and Access Management (IAM) /LLM MCP & agent gateway and auth server with web UI supporting OpenClaw, MCP, OAuth, OIDC, SAML, CAS, LDAP, SCIM, WebAuthn, TOTP, MFA, Face ID, Google Workspace, Azure AD
ghidraMCP
★ 9.0kAn Model Context Protocol server that enables LLMs to autonomously reverse engineer applications by exposing Ghidra's decompilation and analysis tools. It allows AI agents to list code structures, rename methods, and analyze binaries directly through
HexStrike AI
★ 8.9kHexStrike AI MCP Agents is an advanced MCP server that lets AI agents (Claude, GPT, Copilot, etc.) autonomously run 150+ cybersecurity tools for automated pentesting, vulnerability discovery, bug bounty automation, and security research. Seamlessly b
IDA Pro MCP
★ 8.7kEnables AI-assisted reverse engineering in IDA Pro by providing tools to analyze binaries, decompile functions, manage comments, search patterns, and interact with the IDA database through natural language.
Anthropic Cybersecurity Skills
★ 6.6k754 structured cybersecurity skills for AI agents · Mapped to 5 frameworks: MITRE ATT&CK, NIST CSF 2.0, MITRE ATLAS, D3FEND & NIST AI RMF · agentskills.io standard · Works with Claude Code, GitHub Copilot, Codex CLI, Cursor, Gemini CLI & 20+ platform
Hooker
★ 5.1k🔥🔥 hooker is a Frida-based reverse engineering toolkit for Android. It offers a user-friendly CLI, universal scripts, auto hook generation, memory roaming to detect activities/services, one-click SOCKS5 proxy setup, Frida JustTrustMe, and BoringSSL u
Browse More Security MCP Servers
Explore all security servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Agentic Radar in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Agentic Radar?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.