tech
Use Fable Before You Know What to Ask
Plus: A manual for handing recurring work to Opus, a skill for judging new tools in context, and the specialist model that beat the frontier for less
TL;DR
- Fable's strength lies in identifying decisions and premises that users are unaware of.
- It helps differentiate between tasks with execution complexity and those with flawed initial goals or standards.
- "Unknowns" are identified as gaps between the AI's prompt and real-world constraints, categorized as "unknown knowns" and "unknown unknowns."
- Examples show Fable identifying implicit criteria (unknown knowns) and questioning unvalidated goals (unknown unknowns).
- Fable is recommended for tasks where the map (prompt) is incomplete, rather than solely for large-scale jobs.