The practical collection

Good work becomes
something you can use.

Prompts, workflows, skills and guides for working thoughtfully with AI. Read the full material here, borrow what helps, or keep a Markdown copy.

A growing, curated collection connecting the journal’s lessons with reusable practice. Each item states its purpose, audience and evidence status. The current materials are independently written public companions with clearly labelled examples.

For project outcomes and professional context, explore the work and Rachel’s biography. The collection grows as materials are reviewed for public release.

Teaching or coaching? Explore the dedicated educator toolkit for 10 complete workflows and a teacher-owned check-and-repeat log.

Building with AI? Explore Agentic Systems for bounded tool experiments and Workspace Development for clearer context, reusable workflows and evidence-led improvement.

Choose a learning category

30 complete artifacts. Three ways in.

Every counted artifact has its own complete method, worked example, checks, public source notes and Markdown download. Collection hubs, the repeat log and alternate downloads are not extra artifacts.

10 complete artifacts

Educators

Create lessons and materials, calibrate assessment, adapt access and check the teaching decision.

Browse Educators

10 complete artifacts

Agentic Systems

Choose an approach, bound tools and handoffs, rehearse failures and evaluate costs.

Browse Agentic Systems

10 complete artifacts

Workspace Development

Organize knowledge, synthesize research, compare prompts and review a reusable release.

Browse Workspace Development

The next useful step

Grow the learning.
Keep the evidence.

A research-backed expansion plan: strengthen the existing collection, test a small industry-specific practice, and carry useful lessons into the next attempt.

Priority waves, proposed sector opportunities and the checks each release must meet. Planned work is labelled throughout.

Explore the expansion roadmap

Make the next step useful

Find a route by your goal

Choose the result you want to keep, then follow one of nine guided sequences.

Nine learning paths connect the same thirty foundational artifacts. Six synthetic practice cases provide additional rehearsal; they are not counted as new foundational artifacts.

Educators

Build your first trustworthy lesson
One teacher-edited lesson plan with a goal, prerequisites, exact tasks, and a feasible sequence; A learner task and exit check with stable labels, plus a separate independently checked teacher key; A short change record identifying the substantive choices you made and any held material.
150 min · proposed study budget
Practice feedback and scoring with privacy
A checked task, a small criterion rubric, and three wholly invented anchors with teacher reference annotations; A disagreement record containing actual quoted evidence, any abstention, and the teacher’s reason for keeping or rejecting a suggestion; One edited feedback comment, a revision opportunity, a checked fresh task, and a privacy stop decision.
110 min · proposed study budget
Rehearse a collaborative teaching improvement
A short original discussion sequence with evidence-based questions and alternative response routes; Two revision options with a trade-off, plus one teacher-selected and rehearsed move; An optional teacher-owned three-line note and a voluntary reflection point without surveillance or evaluation.
105 min · proposed study budget

Try the 3 synthetic practice cases

Agentic Systems

Build your first bounded experiment
One architecture decision, a permission matrix, a read-only tool contract, and a three-case rehearsal log with an explicit keep/revise/hold decision.
140–180 min · proposed study budget
Carry evidence safely through delegation
A source-to-claim ledger, two restricted assignment packets, checked specialist returns, and a manager decision with a boundary-failure log.
120–160 min · proposed study budget
Decide what survives a provider failure
A dependency map, output-contract check, degraded-response card, and recovery decision with a bounded measurement plan.
90–120 min · proposed study budget

Try the 2 synthetic practice cases

Workspace Development

Turn scattered context into an inspectable pack
A bounded task brief, three-source register, disagreement map, and versioned knowledge pack with answerability tests.
85–115 min · proposed study budget
Keep a bundle consistent when facts change
A candidate fact set, three corrected excerpts, dependency map, issue log, and a draft release manifest with a visible hold or review decision.
100–140 min · proposed study budget
Decide whether a workflow earns reuse
A frozen comparison plan, candidate and baseline records, explicit test outcomes or unrun labels, and a maintenance note with a known-good copy.
150–210 min · proposed study budget

Try the 1 synthetic practice case

Looking for a specific artifact?

Open a category for its complete ten-item index. These are the same foundational methods linked from the paths.

Educators · 10 artifacts

Agentic Systems · 10 artifacts

Workspace Development · 10 artifacts

If you only need one small tool

The five journal companions cover a narrower job: plan a visual, review a printable file, check a claim, keep a learning journal, or choose and adapt a material. They are additional companions, not part of the 30-artifact learning-library count.

A first attempt you can actually finish

Read the required inputs and output before copying a prompt. Use the supplied synthetic example or other material you are permitted to use. Replace placeholders, perform the stated checks and save one result with its version. If a required source or permission is missing, record the gap and stop the dependent step.

Markdown is a plain-text format. You can open a download in a text editor or notes app; formatting symbols such as headings and list markers remain visible. The complete method is also readable on its web page. Neither format configures tools, grants permissions or guarantees that the generated result will be correct.