prompt-optimizerlisted

Rigorous mode prompt optimizer — audits a polished prompt, decides output architecture (single-turn vs. chained multi-stage), and rewrites it to elicit state-of-the-art output from Claude. Triages every input into OUT OF PURVIEW / BORDERLINE / IN PURVIEW; rigorous optimization (architecture decision, 9-dim diagnostic, archetype detection, capability activation) applies ONLY to IN PURVIEW. When chain signals fire, produces a chain deliverable (README + Project orchestrator + per-stage prompts) instead of a single-turn prompt. Simple prompts return substantially unchanged with a triage note. Use when the user asks to optimize, audit, score, diagnose, sharpen, or level up a prompt; trigger phrases include "optimize this prompt", "PO this", "run this through the optimizer", "sharpen this prompt". For messy dictated or flow-state input, a separate flow-mode counterpart exists. Do NOT trigger when the user wants the prompt EXECUTED rather than optimized.
viktor-milev/claude-skills · ★ 0 · AI & Automation · score 72

Install: claude install-skill viktor-milev/claude-skills

# Prompt Optimizer (State-of-the-Art Edition) ## CHANGELOG **v4.1 — May 2026.** Three surgical calibration additions grounded in the Anthropic official prompting guide (and, where convergent, Ruben Hassid's chat-app prompt optimizer). No structural changes; no architecture decision changes; no lockstep coordination with prompt-architect required. - **`<examples>` elevated to first-class structural tool (Step 6 Branch A).** v4's structural tools list included `<role>`, context-first ordering, `<constraints>`, motivated constraints, `<output_format>`, `<failure_modes>`, `<evaluation_criteria>`, domain-specific tags, and `[bracketed placeholders]` — but omitted `<examples>` as a first-class tool, despite Anthropic identifying examples as "one of the most reliable ways to steer Claude's output format, tone, and structure." Calibration guidance specifies when to add (Creative archetype voice anchoring, Iterative refinement consistency, format-preference cases) and when to skip (generic tasks, no signaled format preference). Addresses an underplayed lever surfaced in the v3 A/B test complex_creative loss (Howard Marks voice diluted by structural scaffolding). - **Quote-grounding pattern for long-document tasks (Step 6 Branch A).** Added an explicit structural tool: for prompts loading ≥3K tokens of source material, instruct Claude to extract relevant passages into `<quotes>` tags before reasoning. Per Anthropic's long-context guidance, materially reduces drift and hallucination