Agent Wiz
Open Source
Visualize workflows, detect vulnerabilities, and threat-model your agentic systems — right from your terminal.
Agent-Wiz is an open-source CLI tool by RepelloAI that extracts real agent flows, simulates attacks using MAESTRO, and lets you defend your AI before it breaks.
pip install repello-agent-wiz
REAL-WORLD PROBLEM
Multi-agent systems look smart. But behind the scenes, they’re messy, implicit, and deeply vulnerable.
From tool misuse to prompt poisoning and lateral escalation — your agents are talking to each other, to LLMs, and to tools... without guardrails.
Agent-Wiz makes those flows visible and threat-models what others can’t even parse.
Multi-agent systems look smart. But behind the scenes, they’re messy, implicit, and deeply vulnerable.
From tool misuse to prompt poisoning and lateral escalation — your agents are talking to each other, to LLMs, and to tools... without guardrails.
Agent-Wiz makes those flows visible and threat-models what others can’t even parse.
FEATURES
AST-based flow extraction
AST-based flow extraction
Parses orchestrator files (Python) to generate true runtime agent→tool→LLM graphs
Parses orchestrator files (Python) to generate true runtime agent→tool→LLM graphs
MAESTRO threat modeling
MAESTRO threat modeling
Assesses each interaction across 12 known agentic failure modes
Assesses each interaction across 12 known agentic failure modes
Real-time graph visualization
Real-time graph visualization
Export to .png, .dot, or interactive formats
Export to .png, .dot, or interactive formats
Risk-aware output
Risk-aware output
Get JSON risk summaries, ranked threat paths, and graph-annotated issues
Get JSON risk summaries, ranked threat paths, and graph-annotated issues
Framework agnostic
Framework agnostic
Works out of the box with Autogen, LangGraph, CrewAI, LlamaIndex, OpenAI SDK, n8n
Works out of the box with Autogen, LangGraph, CrewAI, LlamaIndex, OpenAI SDK, n8n
Plug-and-play CLI
Plug-and-play CLI
One command, no config, no vendor lock-in
One command, no config, no vendor lock-in
SECURITY INSIGHTS
Detects privilege escalation vectors across agents
Detects privilege escalation vectors across agents
Highlights implicit flows between memory, tools, and LLMs
Highlights implicit flows between memory, tools, and LLMs
Flags prompt confusion, ghost agents, infinite recursion
Flags prompt confusion, ghost agents, infinite recursion
Outputs MAESTRO-structured reports for risk scoring and mitigation
Outputs MAESTRO-structured reports for risk scoring and mitigation
Because you can’t secure what you can’t see.
Because you can’t secure what you can’t see.
pip install repello-agent-wiz
Sign up for Repello updates
Subscribe to our newsletter to receive the latest insights on AI security, red teaming research, and product updates in your inbox.
8 The Green, Ste A
Dover, DE 19901, United States of America
Follow us on:
© Repello Inc. All rights reserved.
Sign up for Repello updates
Subscribe to our newsletter to receive the latest insights on AI security, red teaming research, and product updates in your inbox.
8 The Green, Ste A
Dover, DE 19901, United States of America
Follow us on:
© Repello Inc. All rights reserved.