Learn Complex Concepts with the 'Fable Prompt'
When trying to learn a difficult concept, asking an AI for a simple definition often leads to information that is quickly forgotten. Amanda Askell, an in-house philosopher at Anthropic, shared a creative solution: the 'Fable Prompt'. This is one of the more unique advanced AI prompting techniques we've seen.
The trick is to ask the AI to explain a graduate-level concept from a specific field by writing a fable that embodies it. The story should be structured so the reader only realizes the concept being illustrated near the end. This narrative approach helps build a more durable intuition for the idea than a dry, technical explanation.
I want you to select a concept at about the graduate-student level from the field of [YOUR FIELD HERE]. Then indirectly explain this concept completely by writing a fable. Structure it so that only toward the very end do readers gradually realize what the concept actually is. After the story, add a section that clearly articulates the concept you just conveyed.A Two-Step Dictation Strategy for Polished Writing
For those who prefer speaking to typing, a structured dictation process can dramatically improve the quality of written documents. This method uses a collaborative editor like Claude Code.
- Draft the Outline: Start by dictating a command to the AI, such as: “Draft an outline for a short internal memo about [topic].”
- Isolate the Draft: Crucially, instruct the agent: “Save this as the initial draft and do not edit it. Now create a separate working draft that we can revise.”
- Review and Comment: Read the working draft and use a dictation tool like Typeless to leave comments, pointing out inaccuracies, missing information, or generic-sounding phrases.
- Refine with a Prompt: Finally, prompt the agent to “Rewrite the draft using the comments I left. Write it in my tone. Use my verbiage. No em dashes. Preserve the core points, but cut anything that sounds generic.”
By regularly comparing the initial draft with the final one, the agent can learn your editorial preferences and improve its first drafts over time.
How to Generate Viral LinkedIn Posts with AI
You can also apply structured prompting to social media content creation. Using a tool like Taplio, which integrates an AI assistant, you can build a repeatable workflow for generating engaging LinkedIn posts.
Start with a detailed prompt to generate ideas that are tailored to your profile and audience. The key is to be specific about the desired tone, format, and goal of the post.
Analyze my LinkedIn profile, posts, and audience, then generate 10 LinkedIn post ideas on using AI at work. Each idea should come from a real work moment, tie AI to leverage or career impact, challenge a common assumption, and match my clear, practical tone. Avoid hype. Optimize for replies over likes. Output only the 10 opening lines or angles.Once you select an idea, ask the AI to turn it into a simple outline, refine the outline yourself, and then have the AI draft the full post. This structured approach ensures the final content is aligned with your strategy and requires only minor edits to personalize the voice.
More Prompting Advice: Skillification and Documentation
Beyond specific prompts, a broader strategic approach can prevent common AI agent failures. Y Combinator CEO Garry Tan advocates for “skillification.” This framework involves forcing an AI to execute precise local scripts for deterministic tasks, like historical calendar lookups, rather than letting it try to solve them in its latent space. This avoids recurring mistakes and wasted compute.
Additionally, good documentation is critical for agent performance. A well-written `AGENTS.md` file with specific, learnable patterns can act as a model upgrade. The documentation should target the specific problems you are trying to solve, as different patterns can move different metrics.
