The practice of decomposing AI-mediated work into verifiable moves and giving each move the minimum sufficient context required to succeed.
Read the Manifesto →What is the minimum sufficient context required for the next verifiable move?
How the methodology emerged from practice. Not from reading about AI workflows — from building, failing, correcting, and capturing what made the difference.
open file →Ten principles for AI-mediated work. Context is the craft. Sufficiency beats volume. The move is the unit of work. Verification precedes trust. Humans own intent.
open file →Move sequencing, decomposition by uncertainty, the Move Card, the three practitioner stances, and how the loop compounds over time.
open file →Move Card template, Sufficiency Check, Verification Contract, Context Ledger, Monday-morning checklist, and the anti-patterns to avoid.
open file →Context is not storage.
Context is instruction, evidence, boundary, and proof.
Extreme Contexting did not begin as a theory. It emerged from building AI-mediated production systems under real pressure — where briefs mattered, validators mattered, sequencing mattered, and accumulated lessons from prior failures mattered more than the model itself.
Its origin is its first proof case: Extreme Contexting was developed by practicing Extreme Contexting.