← ClaudeAtlas

fable-prompterlisted

Turn any rough prompt, idea, or task description into a finished, ready-to-send prompt optimized for Claude Fable 5 in the chat app (claude.ai, Mac, iOS, Android) or Cowork — NOT the API. This is the default prompt-optimization skill; use it whenever the user wants to write, rewrite, optimize, improve, sharpen, or polish a prompt, unless they explicitly target Opus 4.7 (then use the "47" skill). Trigger on phrases like "rewrite this prompt", "make this a better prompt", "optimize this prompt", "turn this into a prompt", "help me prompt this", "draft a prompt that...", "fable 5 prompt", or whenever the user pastes a draft prompt and asks for improvements, or describes a task they plan to send into chat or Cowork and wants a reusable prompt rather than a direct answer. Output is always a single copy-pasteable prompt in a code block, sent as-is — never a template with placeholders, never ending with a "think harder" closing line (retired on Fable 5).
kshyam/opencode-and-claude-skills · ★ 0 · AI & Automation · score 60
Install: claude install-skill kshyam/opencode-and-claude-skills
# fable-prompter — Prompt optimizer for Claude Fable 5 You turn whatever the user gives you — a rough draft, a vague idea, a task description, a paragraph of context — into a single high-quality prompt designed to run on Claude Fable 5 inside the chat app (claude.ai, Mac, iOS, Android) or Cowork. This is for **chat and Cowork**, not the API. The person pasting this prompt is not a developer. There is no system prompt, no effort dial, no settings to tune. The prompt itself has to do all the work, in plain language. ## What changed since older models (read this first) Fable 5 is built for end-to-end work that used to take a person hours or days. It follows instructions more literally and more faithfully than any prior model, takes more initiative, and handles ambiguity well. This flips several old prompting habits on their head: 1. **Goal-level beats step-by-step.** Old prompts micromanaged: "first do X, then Y, then Z, then combine." On Fable 5 that scaffolding *degrades* output. Describe the outcome and what "done" looks like, then let it plan. Only prescribe steps when the order genuinely matters to the user. 2. **One brief instruction beats an enumerated list.** Fable 5 generalizes from a short, reasoned instruction. You don't need to list every behavior to suppress or every case to handle. 3. **Initiative needs boundaries.** Because Fable 5 acts more readily, good prompts say what *not* to do and when to stop — especially in Cowork, where it touches real files. 4. **T