"Because cheating shouldn't be easier than studying."
Argus is a high-performance, real-time network monitoring system engineered for high-stakes exam environments. Operating directly at the network packet level, it eliminates loopholes like VPNs, DNS-bypassing, and clock manipulation with surgical precision.
Developed for the Software Project Lab 1 at the Institute of Information Technology, University of Dhaka.
- DPI & SNI Extraction: Bypasses basic DNS filtering limitations by extracting Server Name Indication (SNI) from initial TLS handshakes before encryption fully takes over.
- VPN Detection: Analyzes payload randomness via Shannon Entropy to identify encrypted tunnels or VPNs without needing to decrypt data.
- Hybrid Cryptography: Secures the communication tunnel between student and supervisor using a robust mix of asymmetric (RSA) and symmetric (RC4/XOR) encryption.
- Precision Anti-Tamper: Employs Cristian’s Algorithm to sync clients with a trusted server time, defeating attempts to manipulate local OS system clocks.
- Multiplexed Dashboard: A single-threaded I/O multiplexing (select()) architecture capable of asynchronously monitoring 100+ students via a sleek ImGui/OpenGL interface.
Argus is built on highly optimized data structures to ensure enforcement happens with zero network latency:
-
Domain Resolution: Trie-based recursive wildcard matching and Polynomial Rolling Hashes for
$O(1)$ lookups. -
Information Theory: VPN detection via the Shannon Entropy formula:
$$H(X) = - \sum_{i=1}^{n} P(x_i) \log_2 P(x_i)$$ - Data Compression: Huffman Coding shrinks violation logs to fit within the strict 512-byte binary protocol limit.
- Supervisor Server (
echo_supervisor.cpp): The command center. Processes handshakes, decrypts telemetry, and provides a real-time dark-themed dashboard. - Student Agent (
echo_student.cpp): A lightweight background sniffer with a watchdog thread that sends immediate tamper alerts if the process is terminated.
Prerequisites: C++17, libpcap, GLFW, OpenGL 3.2+.
Developed for excellence at IIT, University of Dhaka.