Case Study: Paramiko

Repository: paramiko/paramiko Stats: 55 files, 962 functions Drift Score: 0.468 (MEDIUM) | Time: 5.8s

Key Findings

5 God Module Candidates (AVS)

Paramiko's core contains five modules with high coupling that concentrate responsibilities: transport.py (Ca=3, Ce=35), util.py (Ca=22, Ce=8), common.py (Ca=27, Ce=2), pkey.py (Ca=6, Ce=18), and message.py (Ca=17, Ce=5). Of these, transport.py has the highest impact — a single change there ripples across 3 dependent modules.

This is the classic "god module" pattern in long-lived protocol libraries: core transport logic grows over years as features are added, but responsibility boundaries are never split.

18 Circular Import Chains (CIR)

Drift detected 18 circular import chains, the longest involving 6 modules: paramiko → packet → util → hostkeys → pkey → message → paramiko. These cycles indicate tangled module responsibilities and can cause subtle runtime ImportErrors depending on import order.

54 Unexplained High-Complexity Functions (EDS)

The Explainability Deficit Signal flagged 54 functions with high cyclomatic complexity but no docstrings. The top three are Transport.run, SSHClient.connect, and BufferedFile.readline — all critical protocol-handling functions.

Interpretation

Paramiko is a mature, stable SSH library — but stability has come at the cost of architectural clarity. The god-module pattern in transport.py and the circular import chains create a high blast radius for changes and make contributor onboarding difficult.

Recommendation: Start with transport.py — extract protocol state management into a separate module to reduce coupling. For the circular imports, introduce a shared types module and use TYPE_CHECKING-guarded imports to break the longest chains first.