Claude Fable 5 is the most powerful model there is, and most people never know when to actually use it. This is the exact skill that makes that call. Copy the block below, give it to your Claude, and the skill builds itself. Point it at anything and it tells you if Fable 5 is worth it and what to do with it.
← All Claude SkillsThis is the whole thing. You do not need to understand it or fill anything in. Fable 5 is your most powerful model, and its value comes from pointing it at the right work. This skill makes that call for you and hands back the plays worth it. Copy the two blocks below and give them to your Claude.
Step 1. Copy the skill block below.
Step 2. Open Claude and paste this instruction first, then paste the block underneath it:
Turn the block below into a Claude skill and install it. Create the skill file exactly as written. Then personalize it with smart defaults from what you know about me: my business, my voice, my current priorities, and where I keep my work. Confirm when it is installed and tell me how to run it.
--- name: fable-lens description: Run anything I'm working on through the Fable 5 discernment filter and surface the highest-leverage plays. Use whenever I say "fable lens this," "run the fable lens," "is this fable worthy," "fable play for [topic]," or any time I want to create something with Claude Fable 5. Also suggest it when I'm about to point Fable at routine production work. --- # The Fable Lens The standard for how I create with Claude Fable 5. This skill holds the discernment so I never have to remember it. Point it at anything, get the best Fable plays back. ## Core Principle Fable writes the assets, cheaper models run them forever. Fable time goes to durable intelligence. Routine production stays on cheaper, faster models. ## The Fable-Worthiness Test (both must be true) 1. It needs frontier intelligence. A cheaper model would produce a meaningfully worse result. 2. The output compounds. It gets used repeatedly, or the decision it informs is worth six figures plus. ## What Fable Is For (the five categories) 1. Architecture, not tasks. Systems that run thousands of times after one design session. 2. Irreversible decisions. Org charts, engine selection, naming, schema. High regret if wrong. 3. IP creation. Frameworks, skills, and assets that become products and client deliverables. 4. Full-context synthesis. Reading the whole business at once and finding what no single conversation reveals. 5. Deep builds. Full systems, sites, and agentic networks an expert would need a week for, shipped in a session. ## Operating Principles On Every Play - 92/8: agents and AI do the work, I make the judgment calls. - 10-80-10: I set the vision, co-creation in the middle, I review before ship. - Definition of Done on everything, one or zero. Did it ship or not. - Ground Rule: every Fable session ends with output written to a file or my second brain. Nothing lives only in chat. ## Route To A Cheaper Model Instead - Routine production and anything an existing skill or workflow already handles. - Output that gets consumed once and disappears. ## Process Step 1 — Intake. Take whatever I bring. Project, idea, transcript, frustration, goal. No format required. Step 2 — Sweep. Look at it across my whole business, since the best play may live outside the area I raised. Step 3 — Plays. Present 3-5 ranked plays. Each gets: The Play (named), Category (which of the five), Why Fable (what frontier intelligence adds), Durable Asset (what runs it forever), Revenue Connection, Definition of Done (one or zero), and First Prompt (the opener I hand Fable). Step 4 — I rule. Present the ranked plays. I pick. One word is enough. Step 5 — Session spec. For the chosen play: outcome, sources to read, where the output gets saved, Definition of Done. Paste-ready. ## Quality Standards - Every play passes the Worthiness Test or it stays off the list. - "Nothing here is Fable-worthy" is a valid answer and a win. Name the cheaper model that should handle it instead. - Plays wire into infrastructure I already own. No new shells. - Living check: before generating new plays, flag any recent Fable-built asset not yet in real use. Wiring it in competes for the top slot.
Claude reads the block, builds the skill file, and installs it. From then on, you just run the skill in any conversation. No folders to find, no code to write.
Grow your business with AI, made simple and fun. Like having a Chief AI Officer by your side.
Here's the shape of what this skill hands you when it runs. Copy it, use it, keep moving.
You don't need to hold any of this in your head. That's the skill's job. If you're curious what it's doing for you, here's the idea underneath it.
Think of Fable 5 like hiring the most expensive expert in the world for an hour. You do not have them answer email. You have them design the thing that runs a thousand times after they leave.
Fable writes the assets. Cheaper models run them forever. The skill spots which work is worth the expert, so you point your best model at the foundation and let everything else run cheap.
When work is worth Fable, it almost always falls into one of these five. This is what the skill scans for. It is a useful picture of where your best model earns its keep.
Systems that run thousands of times after one design session. One hour designing the right structure pays out every day forever.
Naming, structure, which direction to commit to. The calls that are expensive to undo. Worth the smartest model in the room.
Frameworks, skills, and assets that become products and client deliverables. Build it once, it becomes the thing you sell.
Finding what no single conversation reveals. You cannot hold that much context in your head. Fable can.
Full systems, sites, and agent networks that would take an expert a week, shipped in a single session. This is where the leverage lives.
The two-part test the Fable Lens runs on every piece of work. First, it needs frontier intelligence, where a cheaper model would produce a meaningfully worse result. Second, the output compounds, meaning it gets used again and again or informs a decision worth six figures or more. Work passes both parts to earn your best model, or the skill routes it to a cheaper, faster one.
The Fable Lens looks for five categories: architecture that runs thousands of times after one design session, irreversible decisions like naming and structure, creating your own frameworks and IP, reading your whole business at once, and deep builds an expert would need a week for. Routine production that gets used once stays on a cheaper model.
You point it at a project, an idea, or a goal, and it hands back three to five ranked plays worth your best model. Each play names the category, what frontier intelligence adds, the durable asset it produces, the revenue connection, a one-or-zero Definition of Done, and the First Prompt to start. If nothing qualifies, it names the cheaper model to use.