agentic-radar
A security scanner for your LLM agentic workflows
claude mcp add --transport stdio splx-ai-agentic-radar python -m agentic_radar
How to use
Agentic Radar is a security analysis tool for agentic workflows. It scans systems to produce a comprehensive HTML report that visualizes agentic workflows, identifies tools in use (external and custom), maps detected MCP servers, and correlates potential vulnerabilities with known security frameworks. The CLI provides two core commands: scan and test. Use scan to analyze a project’s workflow graph and generate a report, for example by pointing the tool at a language graph, CrewAI workflows, or OpenAI Agents assets. Use test to run vulnerability assessments on an agentic workflow, which requires an OpenAI API key to run the checks across the selected framework. Advanced usage includes enabling optional extras (like CrewAI or OpenAI Agents integrations) for deeper tool descriptions and richer vulnerability mappings.
How to install
Prerequisites:
- Python 3.8+ and pip installed on your system
Install Agentic Radar:
pip install agentic-radar
Verify installation:
agentic-radar --version
Optional advanced installations:
- CrewAI enhancements (requires CrewAI support in your environment):
pip install "agentic-radar[crewai]" - OpenAI Agents enhancements (for richer OpenAI Agents integration):
pip install "agentic-radar[openai-agents]"
Usage after installation is ready via the CLI as described in the next section.
Additional notes
Notes and tips:
- To run vulnerability tests, you must set OPENAI_API_KEY in your environment.
- Some features depend on external tool integrations (e.g., CrewAI, OpenAI Agents). Install the corresponding extras if you plan to use those features.
- The MCP server setup in this repo uses the Python package entry point; ensure your Python environment has access to the agentic-radar executable in your PATH when invoking the MCP server via MCP tooling.
- If you encounter permission errors on installation, consider using a virtual environment (venv) to isolate dependencies.
- The tool generates an HTML report that you can view in a browser; for long-running scans, ensure your environment supports the required resources for graph rendering and vulnerability mapping.
Related MCP Servers
dify
Production-ready platform for agentic workflow development.
learn-ai-engineering
Learn AI and LLMs from scratch using free resources
solace-agent-mesh
An event-driven framework designed to build and orchestrate multi-agent AI systems. It enables seamless integration of AI agents with real-world data sources and systems, facilitating complex, multi-step workflows.
MassGen
🚀 MassGen is an open-source multi-agent scaling system that runs in your terminal, autonomously orchestrating frontier models and agents to collaborate, reason, and produce high-quality results. | Join us on Discord: discord.massgen.ai
mcp-client-for-ollama
A text-based user interface (TUI) client for interacting with MCP servers using Ollama. Features include agent mode, multi-server, model switching, streaming responses, tool management, human-in-the-loop, thinking mode, model params config, MCP prompts, custom system prompt and saved preferences. Built for developers working with local LLMs.
zypher-agent
A minimal yet powerful framework for creating AI agents with full control over tools, providers, and execution flow.