clarifylisted

# Clarify - Intensive Questioning (v2.37) Systematically gather requirements using **TLDR semantic search** + AskUserQuestion tool. ## v2.88 Key Changes (MODEL-AGNOSTIC) - **Model-agnostic**: Uses model configured in `~/.claude/settings.json` or CLI/env vars - **No flags required**: Works with the configured default model - **Flexible**: Works with GLM-5, Claude, Minimax, or any configured model - **Settings-driven**: Model selection via `ANTHROPIC_DEFAULT_*_MODEL` env vars ## Quick Start ```bash /clarify # Start intensive questioning for current task ``` ## Pre-Clarification: TLDR Semantic Search (v2.37) **AUTOMATIC** - Before asking questions, use semantic search to understand existing code: ```bash # Find existing related functionality (95% token savings) tldr semantic "$USER_TASK_KEYWORDS" . # Example: For "add authentication", find existing auth code tldr semantic "authentication login session user password" . # Get structure overview for context tldr structure . --lang "$PRIMARY_LANGUAGE" ``` This helps formulate better questions based on what already exists in the codebase. ## Aristotle-First Clarification (v3.0) Before asking structured questions, apply Aristotle Phase 1 (Assumption Autopsy): 1. **What assumptions are embedded in the user's request?** Identify inherited framing. 2. **What clarifications challenge assumptions vs confirm them?** Prioritize assumption-challenging questions. 3. **What would change if the core assumption is wrong?** This ide