The Shelf

The program's reference page. Templates, rules of thumb, and plain-English definitions. No theory, just the tools.

The starter standing instructions

Five rules, each one earned in a lesson. Copy them, rewrite every rule in your own words (Lesson 7 explains why), and put them wherever your AI tool keeps standing instructions: a system prompt, custom instructions, or a profile.

RULES FOR WORKING WITH ME

1. Never tell me something is done unless you can show me
   evidence: a record, an ID, a change I can verify.

2. If you are not sure, say you are not sure. A wrong answer
   costs me more than a slow one.

3. Before we build around any constraint, check it. Tell me
   what you verified and how.

4. Every fact lives in exactly one place. Tell me when you
   see the same fact in two places.

5. When something breaks, give me two or three possible
   causes before any fix.

Rules of thumb

Evidence over reasoning.
If you can't verify it, it didn't happen.
Confidence is a writing style. Evidence is a thing in the world.
The same fact in two places is a bug.
Chat is a dead end you can live in for years.
Before you build around a wall, touch it.
Your first guess is a prediction machine too, and it is overconfident too.
The system is working when it needs you less.
Never delegate a task you cannot describe the finished state of.
Trust a test, not an impression.
Match the weight of the tool, and of your attention, to the weight of the consequence.
Parallel only what is independent. Merge at one point you control.

Plain-English glossary

Reading a definition and being able to say it in your own words are two different things. The cards check which one you have.

What mature doctrine looks like

A glimpse of the destination. After a year or two of real use, a working set of standing instructions grows from five rules into something like this (excerpts from real ones, lightly edited):

Define what "done" means before starting, not after.

Break big work into pieces that can each be checked
on their own. One piece, one risk, one clear finish line.

When the machine fails twice the same way, stop prompting
harder. Change what it sees: rules, examples, data.

Review the work like an editor, not a fan. The machine
does most of the doing; you own the judgment.

Track what failed and why. The log of failures is the
curriculum for the next month's rules.

Where to go next

The next step is not more reading.

Do the practices for real. Especially the Walls tally (Lesson 4), the standing-instructions week (Lesson 7), and the kept audition (Lesson 10). Reading them counts for nothing, doing them counts for everything.
Answer the return questions when they find you. They are how this program makes things stick.
Run one real piece of work through the full toolkit. Define done, delegate it, verify the result, log what failed. That run, not the reading, is the graduation.
Then teach the material to someone else. Gaps in your explanation are the most honest map of what to revisit.