python-explorelisted

# Python Explore Build a mental model with pyeye before touching Python code. ## Skill Type: Mixed Rigid/Flexible **Rigid gates (non-negotiable):** - Always produce a written summary before proceeding - Always use pyeye before Read() on unfamiliar code - Never re-explore what's already in context **Flexible path (use judgement):** - Depth scaling based on task scope - Which specific pyeye calls to make - When the mental model is "sufficient" ## Do NOT Trigger When - User says "show me", "print", "display", "read this file" — respect explicit requests - User says "run", "execute" — not exploration - User asks "what's the syntax for" — language question, not codebase question - Single-line typo/string fix where user gives exact location - Adding a new test case (not modifying test infrastructure) - Pyeye output for this module is already visible in conversation context If pyeye output is already in context, skip to summary and proceed. ## Depth Scaling | Task Scope | Pyeye Calls | |------------|-------------| | Single class/function change | `lookup()` (primary entry point) — for advanced use: `find_references()` with field filtering | | Cross-module change | `lookup()` + `analyze_dependencies()` | | "I don't know where to start" | Full: `list_project_structure()` → `list_modules()` → drill down | Never run `list_project_structure()` for a scoped single-symbol task. ## Relationship Queries Require Pyeye, Not Grep **These patterns start with `lookup()`. Use `find_r