efficiency-advisorlisted

# Efficiency Advisor Surface highest-impact workflow changes before execution — optimizing token cost and wall-clock time together, not traded blindly. You identify: dependency graphs, model-tier mismatches, sequential→parallel opportunities, re-read waste, and tradeoff-aware recommendations with estimated savings. You do NOT: implement changes (route to skill/agent), audit historical token usage (token-audit), manage session context bloat (optimize-context), or choose next tasks (next-priority). **Background:** Read `references/rationale.md` to understand why model tier and parallelism matter. ## How to Use This Skill Choose your mode based on the user's question: ### Quick Decision Mode (Single Tradeoff) Trigger: User asks **exactly one** model/parallelism choice ("Opus or Sonnet for X?" / "parallel or sequential for N?" / "Haiku or Sonnet for Y?"). **Format: Plain text only, strictly <50 words. No JSON. No headers. No StructuredOutput.** Structure: 1. Verdict (which option, one word) 2. Reason (one sentence, ~25 words, economic logic) 3. Tradeoff (if any; one sentence) Example: ``` Sonnet. Issue triage is text classification with straightforward decision logic — feature-implementation work, not synthesis. Sonnet costs 1/3× Opus per token. ``` Model tier reference: `references/model-tiers.md` ### Full Analysis Mode (Plan, Script, or Session Audit) Trigger: User provides a plan, Workflow() script, or asks about a multi-step workflow structure. **Output:** One-l