PANTHER is a plugin‑based, research‑grade test harness that lets you design, reproduce, and analyse complex network‑protocol experiments without hand‑rolling scripts or bespoke infrastructure.
!!! info "What PANTHER Solves" - Protocol Validation: Test QUIC or custom protocol implementations under failure, jitter, or adverse timing - Performance Profiling: Analyze CPU, heap, and syscall characteristics across different builds or OS kernels - Formal Verification: Run conformance checks (Ivy) in deterministic network simulation (Shadow)
Core characteristics:
▸ Reproducible: every experiment is defined in a single YAML file and executed in an isolated container environment.
▸ Extensible: a plugin system adds new protocols, services, profilers, or network back‑ends with minimal boilerplate.
▸ Multi‑audience: useful to academic researchers, industrial developers, security analysts, SRE teams, and educators.esting Harness for Extensible Research
PANTHER experiments follow a 4-phase execution model:
- Load configurations and validate experiment setup
- Initialize plugin system and service managers
- Create test case instances
- Discover and load protocol/implementation plugins
- Create service managers for each IUT (Implementation Under Test)
- Generate deployment and execution commands
- Setup network environment (Docker Compose, localhost, or Shadow NS)
- Build container images for protocol implementations
- Deploy services with proper networking and monitoring
- Start services and execute test scenarios
- Monitor execution with automatic packet capture
- Collect results, logs, and performance metrics
- Teardown environment and generate reports
Key Features:
- Reproducible: Every experiment defined in single YAML configuration
- Containerized: Isolated execution environments with Docker
- Event-driven: Real-time monitoring and coordination
- Extensible: Plugin architecture for new protocols and environments
For detailed workflow documentation, see WORKFLOW.md.
| Component | Minimum | Notes |
|---|---|---|
| Python | 3.10 | Use venv for isolation for main functionality. |
| Docker | 27.x | Required for all orchestration modes. |
pyproject.toml is the source of truth for Python dependencies.
requirements.txt is a frozen snapshot—do not edit.
Note: We propose to install slim in our builder, fasten container size.
- Fast-Fail System - Intelligent experiment termination and error handling
- Experiment Reporting - Automatic generation of experiment reports and status summaries
- Configuration Management - Advanced configuration validation, auto-fixing, and protocol-aware port management
PANTHER's extensible plugin architecture enables seamless integration of new protocols, implementations, testing frameworks, and environments. The modern inheritance-based system provides 47.2% average code reduction while ensuring consistent behavior across all plugin types.
Core Plugin Categories:
- Overview - Architecture and design patterns
- Inventory - Complete plugin catalog
- Environment Plugins - Network simulation and execution environments
- Overview - Environment plugin architecture
- Network Environment - Docker Compose, localhost, Shadow NS
- Execution Environment - Performance profiling and analysis
- Protocol Plugins - Protocol definitions and behavioral specifications
- Overview - Protocol plugin patterns
- Client-Server Protocols - QUIC, HTTP, TCP/UDP variants
- Peer-to-Peer Protocols - BitTorrent, WebRTC protocols
- Service Plugins - Implementation testing and verification services
- Overview - Service plugin architecture
- Implementation Under Tests (IUTs) - picoquic, aioquic, quiche, quinn
- Testing Services - Ivy formal verification, custom testers
Plugin System Features:
- Automatic Discovery: Decorator-based registration with metadata validation
- Dependency Management: Semantic versioning and automatic dependency resolution
- Version Configurations: Protocol version-specific configurations (RFC9000, draft-29, etc.)
- Event Integration: Built-in event emission and coordination across plugin lifecycle
- Configuration Schema: JSON Schema-based validation with auto-fixing capabilities
- Performance Optimization: Class caching and lazy loading for improved startup time
- Contributing
- Plugin Development
- Overview
- Environment Plugins
- Protocol Plugins
- Service Plugins
For detailed information on using PANTHER, see the:
Contributions are welcome! To get started:
- Fork the repository.
- Create a new branch for your feature or bug fix.
- Submit a pull request with a clear description of your changes.
For more details, see the Contribution Guide.
For support or inquiries, please contact:
- ElNiak
- Open an issue on the GitHub repository.
For further reading and context on the topics and methodologies used in this tool, refer to the following articles:
- Crochet, C., Aoga, J., & Legay, A. (2024). Formally Discovering and Reproducing Network Protocols Vulnerabilities (NordSec24).
@techreport{crochet2024formally,
title={Formally Discovering and Reproducing Network Protocols Vulnerabilities},
author={Crochet, Christophe and Aoga, John and Legay, Axel},
year={2024}
url={https://dial.uclouvain.be/pr/boreal/object/boreal:292503}
}
- Rousseaux, T., Crochet, C., Aoga, J., Legay, A. (2024). Network Simulator-Centric Compositional Testing. In: Castiglioni, V., Francalanza, A. (eds) Formal Techniques for Distributed Objects, Components, and Systems. FORTE 2024. Lecture Notes in Computer Science, vol 14678. Springer, Cham. https://doi.org/10.1007/978-3-031-62645-6_10
@inproceedings{rousseaux2024network,
title={Network Simulator-Centric Compositional Testing},
author={Rousseaux, Tom and Crochet, Christophe and Aoga, John and Legay, Axel},
booktitle={International Conference on Formal Techniques for Distributed Objects, Components, and Systems},
pages={177--196},
year={2024},
organization={Springer},
doi={https://doi.org/10.1007/978-3-031-62645-6_10}
}
- Crochet, C., Rousseaux, T., Piraux, M., Sambon, J.-F., & Legay, A. (2021). Verifying quic implementations using ivy. In Proceedings of the 2021 Workshop on Evolution, Performance and Interoperability of QUIC. DOI
@inproceedings{crochet2021verifying,
title={Verifying QUIC implementations using Ivy},
author={Crochet, Christophe and Rousseaux, Tom and Piraux, Maxime and Sambon, Jean-Fran{\c{c}}ois and Legay, Axel},
booktitle={Proceedings of the 2021 Workshop on Evolution, Performance and Interoperability of QUIC},
pages={35--41},
year={2021},
url={https://dl.acm.org/doi/abs/10.1145/3488660.3493803}
}
- Crochet, C., & Sambon, J.-F. (2021). Towards verification of QUIC and its extensions. (Master's thesis, UCL - Ecole polytechnique de Louvain). Available at UCLouvain. Keywords: QUIC, Formal Verification, RFC, IETF, Specification, Ivy, Network.
@article{crochettowards,
title={Towards verification of QUIC and its extensions},
author={Crochet, Christophe and Sambon, Jean-Fran{\c{c}}ois}
year={2021},
url={https://dial.uclouvain.be/downloader/downloader.php?pid=thesis%3A30559&datastream=PDF_01&cover=cover-mem}
}
For other useful resources, see the following:
-
McMillan, K. L., & Padon, O. (2018). Deductive Verification in Decidable Fragments with Ivy. In A. Podelski (Ed.), Static Analysis - 25th International Symposium, SAS 2018, Freiburg, Germany, August 29-31, 2018, Proceedings (pp. 43–55). Springer. DOI - PDF
-
Taube, M., Losa, G., McMillan, K. L., Padon, O., Sagiv, M., Shoham, S., Wilcox, J. R., & Woos, D. (2018). Modularity for decidability of deductive verification with applications to distributed systems. In Proceedings of the 39th ACM SIGPLAN Conference on Programming Language Design and Implementation, PLDI 2018, Philadelphia, PA, USA, June 18-22, 2018 (pp. 662–677). ACM. DOI
-
Padon, O., Hoenicke, J., McMillan, K. L., Podelski, A., Sagiv, M., & Shoham, S. (2018). Temporal Prophecy for Proving Temporal Properties of Infinite-State Systems. In 2018 Formal Methods in Computer Aided Design, FMCAD 2018, Austin, TX, USA, October 30 - November 2, 2018 (pp. 1–11). IEEE. DOI - PDF
-
Padon, O., McMillan, K. L., Panda, A., Sagiv, M., & Shoham, S. (2016). Ivy: safety verification by interactive generalization. In Proceedings of the 37th ACM SIGPLAN Conference on Programming Language Design and Implementation, PLDI 2016, Santa Barbara, CA, USA, June 13-17, 2016 (pp. 614–630). ACM. DOI
-
McMillan, K. L. (2016). Modular specification and verification of a cache-coherent interface. In 2016 Formal Methods in Computer-Aided Design, FMCAD 2016, Mountain View, CA, USA, October 3-6, 2016 (pp. 109–116). DOI
-
McMillan, K. L., & Zuck, L. D. (2019). Formal specification and testing of QUIC. In Proceedings of ACM Special Interest Group on Data Communication (SIGCOMM’19). ACM. Note: to appear. PDF