https://github.com/qualifire-dev/rogue
AI Agent Evaluator & Red Team Platform
https://github.com/qualifire-dev/rogue
agents ai ai-agents e2e-testing llm testing testing-framework
Last synced: 9 days ago
JSON representation
AI Agent Evaluator & Red Team Platform
- Host: GitHub
- URL: https://github.com/qualifire-dev/rogue
- Owner: qualifire-dev
- License: other
- Created: 2025-06-05T21:16:31.000Z (11 months ago)
- Default Branch: main
- Last Pushed: 2026-02-24T09:57:08.000Z (2 months ago)
- Last Synced: 2026-02-24T15:44:45.575Z (2 months ago)
- Topics: agents, ai, ai-agents, e2e-testing, llm, testing, testing-framework
- Language: Python
- Homepage: https://qualifire.ai
- Size: 40.7 MB
- Stars: 1,008
- Watchers: 7
- Forks: 158
- Open Issues: 45
-
Metadata Files:
- Readme: README.md
- Contributing: CONTRIBUTING.md
- License: LICENSE.md
- Code of conduct: CODE_OF_CONDUCT.md
- Codeowners: .github/CODEOWNERS
- Security: SECURITY.md
- Agents: AGENTS.md
Awesome Lists containing this project
- awesome-production-agentic-systems - Rogue - dev/rogue.svg?cacheSeconds=86400) - Rogue is an AI agent evaluator and red team platform for testing agents against business policies and security vulnerabilities. (Agent Security)
- awesome-ai-offensive-security - rogue - Tool to stress-test your AI agents before attackers do by finding prompt injection, sensitive data exposure, and excessive agency. (AI Red Teaming (Testing AI Targets))
README
# Rogue — AI Agent Evaluator & Red Team Platform
**Stress-test your AI agents before attackers do.**
[Discord Community](https://discord.gg/EUfAt7ZDeK) · [Quick Start](#-quick-start) · [Documentation](./docs/)
---
## Two Ways to Harden Your Agent
### 🎯 Automatic Evaluation
Test your agent against **business policies** and expected behaviors.
- Define scenarios & expected outcomes
- Verify compliance with business rules
- Watch live conversations as Rogue probes your agent
- Get detailed pass/fail reports with reasoning
**Best for:** Regression testing, behavior validation, policy compliance
### 🔴 Red Teaming
Simulate **adversarial attacks** to find security vulnerabilities.
- 75+ vulnerabilities across 12 security categories
- 20 attack techniques (encoding, social engineering, injection)
- CVSS-based risk scoring
- 8 compliance frameworks (OWASP, MITRE, NIST, GDPR, EU AI Act)
**Best for:** Security audits, penetration testing, compliance reporting
---
## Architecture
Rogue operates on a **client-server architecture** with multiple interfaces:
| Component | Description |
| ---------- | ------------------------------------------- |
| **Server** | Core evaluation & red team logic |
| **TUI** | Modern terminal interface (Go + Bubble Tea) |
| **CLI** | Non-interactive mode for CI/CD pipelines |
https://github.com/user-attachments/assets/b5c04772-6916-4aab-825b-6a7476d77787
### Supported Protocols
| Protocol | Transport | Description |
| ---------- | -------------------- | ---------------------------------------------------------------------------------- |
| **A2A** | HTTP | [Google's Agent-to-Agent](https://a2a-protocol.org/latest/) protocol |
| **MCP** | SSE, STREAMABLE_HTTP | [Model Context Protocol](https://modelcontextprotocol.io/) via `send_message` tool |
| **Python** | — | Direct Python function calls (no network protocol) |
See examples in [`examples/`](./examples/) for reference implementations.
#### Python Entrypoint
For agents implemented as Python functions without A2A or MCP:
1. Create a Python file with a `call_agent` function:
```python
def call_agent(messages: list[dict]) -> str:
"""
Process conversation and return response.
Args:
messages: List of {"role": "user"|"assistant", "content": "..."}
Returns:
Agent's response as a string
"""
# Your agent logic here
latest = messages[-1]["content"]
return f"Response to: {latest}"
```
2. Run Rogue with Python protocol:
```bash
uvx rogue-ai cli \
--protocol python \
--python-entrypoint-file ./my_agent.py \
--judge-llm openai/gpt-4o-mini
```
Or via TUI: select "Python" as the protocol and enter the file path.
See [`examples/python_entrypoint_stub.py`](./examples/python_entrypoint_stub.py) for a complete example.
---
## 🔥 Quick Start
### Prerequisites
- `uvx` — [Install uv](https://docs.astral.sh/uv/getting-started/installation/)
- Python 3.10+
- LLM API key (OpenAI, Anthropic, or Google)
### Installation
```bash
# TUI (recommended)
uvx rogue-ai
# CLI / CI/CD
uvx rogue-ai cli
```
### Try It With the Example Agent
```bash
# All-in-one: starts both Rogue and a sample T-shirt store agent
uvx rogue-ai --example=tshirt_store
```
Configure in the UI:
- **Agent URL**: `http://localhost:10001`
- **Mode**: Choose `Automatic Evaluation` or `Red Teaming`
---
## Running Modes
| Mode | Command | Description |
| ------- | --------------------- | ----------------------- |
| Default | `uvx rogue-ai` | Server + TUI |
| Server | `uvx rogue-ai server` | Backend only |
| TUI | `uvx rogue-ai tui` | Terminal client |
| CLI | `uvx rogue-ai cli` | Non-interactive (CI/CD) |
### Server Options
```bash
uvx rogue-ai server --host 0.0.0.0 --port 8000 --debug
```
### CLI Options
```bash
uvx rogue-ai cli \
--evaluated-agent-url http://localhost:10001 \
--judge-llm openai/gpt-4o-mini \
--business-context-file ./.rogue/business_context.md
```
| Option | Description |
| ------------------------ | ------------------------------------------- |
| `--config-file` | Path to config JSON |
| `--evaluated-agent-url` | Agent endpoint (required) |
| `--judge-llm` | LLM for evaluation (required) |
| `--business-context` | Context string or `--business-context-file` |
| `--input-scenarios-file` | Scenarios JSON |
| `--output-report-file` | Report output path |
| `--deep-test-mode` | Extended testing |
---
## Red Teaming
### Scan Types
| Type | Vulnerabilities | Attacks | Time |
| ---------- | --------------- | ------------- | ---------- |
| **Basic** | 5 curated | 6 | ~2-3 min |
| **Full** | 75+ | 40+ | ~30-45 min |
| **Custom** | User-selected | User-selected | Varies |
### Compliance Frameworks
- **OWASP LLM Top 10** — Prompt injection, sensitive data exposure, excessive agency
- **MITRE ATLAS** — Adversarial threat landscape for AI systems
- **NIST AI RMF** — AI risk management framework
- **ISO/IEC 42001** — AI management system standard
- **EU AI Act** — European AI regulation compliance
- **GDPR** — Data protection requirements
- **OWASP API Top 10** — API security best practices
### Attack Categories
| Category | Examples |
| ------------------ | --------------------------------------- |
| Encoding | Base64, ROT13, Leetspeak |
| Social Engineering | Roleplay, trust building |
| Injection | Prompt injection, SQL injection |
| Semantic | Goal redirection, context poisoning |
| Technical | Gray-box probing, permission escalation |
### Risk Scoring (CVSS-based)
Each vulnerability receives a **0-10 risk score** based on:
- **Impact** — Severity if exploited
- **Exploitability** — Success rate likelihood
- **Human Factor** — Manual exploitation potential
- **Complexity** — Attack difficulty
### Reproducible Scans
```bash
# Use random seeds for reproducible results
uvx rogue-ai cli --random-seed 42
```
Perfect for regression testing and validating security fixes.
---
## Configuration
### Environment Variables
```bash
OPENAI_API_KEY="sk-..."
ANTHROPIC_API_KEY="sk-..."
GOOGLE_API_KEY="..."
```
### Config File (`.rogue/user_config.json`)
```json
{
"evaluated_agent_url": "http://localhost:10001",
"judge_llm": "openai/gpt-4o-mini"
}
```
---
## Key Features
| Feature | Description |
| ------------------------ | -------------------------------------------- |
| 🔄 Dynamic Scenarios | Auto-generate tests from business context |
| 👀 Live Monitoring | Watch agent conversations in real-time |
| 📊 Comprehensive Reports | Markdown, CSV, JSON exports |
| 🔍 Multi-Faceted Testing | Policy compliance + security vulnerabilities |
| 🤖 Model Support | OpenAI, Anthropic, Google (via LiteLLM) |
| 🛡️ CVSS Scoring | Industry-standard risk assessment |
| 🔁 Reproducible | Deterministic scans with random seeds |
---
## Documentation
- **[Quick Reference](./docs/QUICK_REFERENCE.md)** — One-page cheat sheet
- **[Red Team Workflow](./docs/RED_TEAM_WORKFLOW.md)** — Technical deep-dive
- **[Implementation Status](./docs/IMPLEMENTATION_STATUS.md)** — Feature breakdown
- **[Attack Mapping](./docs/ATTACK_VULNERABILITY_MAPPING.md)** — Vulnerability coverage
---
## Contributing
1. Fork the repository
2. Create a branch (`git checkout -b feature/amazing-feature`)
3. Commit changes (`git commit -m 'Add amazing feature'`)
4. Push (`git push origin feature/amazing-feature`)
5. Open a Pull Request
---
## License
Licensed under a proprietary license — see [LICENSE](LICENSE.md).
Free for personal and internal use. Commercial hosting requires licensing.
Contact: `hello@rogue.security`