Plan a Week of Lessons from High-Quality Instructional Materials

The Sunday-night planning session is one of the quiet taxes of teaching. You know the standard you're on. You know roughly where the class is. What's missing is the two or three hours it takes to turn that into five lessons with warm-ups, worked examples, guided practice, and something students can actually hold in their hands the next morning. Multiply by every subject you teach, every week of the year, and you have the reason so many good lessons never get written down.

Claude for Teachers is built to attack exactly that gap. Its named "plan a lesson" workflow doesn't hand you a generic worksheet generator. It asks Claude to plan from high-quality instructional materials, aligned to your state's standards — and then gives you back a draft plan plus the student-facing materials to go with it. This piece walks the workflow end to end, because the difference between a toy and a tool is entirely in how you drive it.

What "plan a lesson" actually does

Start with what happens under the hood, because it changes how you prompt. When you ask Claude for Teachers to plan a lesson, it isn't improvising from whatever it half-remembers about your topic. It's grounded in two things at once. First, widely used curricula that have been mapped to state standards — the launch brings in trusted sources like OpenSciEd for science and Illustrative Mathematics' IM v.360 for math. Second, the Learning Commons Knowledge Graph connector, included for every verified teacher, which gives Claude the academic standards for all 50 states and the fine-grained learning competencies beneath each standard, in the order students typically learn them.

That second layer is the one people underrate. A standard is a destination. The progression is the road to it — the smaller skills a student has to pick up first, and the sequence that makes them stick. Because Claude has both when it plans, the lesson comes out scaffolded and standards-aligned by construction, not retrofitted after the fact. You aren't asking a chatbot to guess at rigor. You're asking a planning partner that already knows what the standard is built from.

The open-source k12-lesson-planning skill is the machinery behind this. It drafts a full plan and the student-facing materials to revise, drawing on those curricula and progressions, with per-subject reference guidance for ELA, math, science, and social studies. If you want the internals, see the companion piece Inside k12-lesson-planning. For the workflow itself, you don't need to touch any of it — inside Claude for Teachers, the skill is already installed.

The prompt that gets a usable draft

The mistake most teachers make on day one is asking for too little. "Make me a lesson on photosynthesis" gets you a lesson on photosynthesis — for nobody in particular. Claude plans better when you plan out loud. Give it the shape of your actual week:

• Name the standard, not just the topic. "6th grade, next unit in our IM v.360 sequence" beats "ratios." Claude can then pull the right progression and align to where you genuinely are.
• Say what already happened. "They just finished the diagnostic and most missed equivalent ratios" tells Claude what to reteach versus what to move past. Context you'd give a student teacher, give Claude.
• Set the constraints. Class length, whether it's a block or a single period, materials you have, materials you don't. A plan that assumes a document camera you lack is a plan you rewrite.
• Ask for the whole arc, then the pieces. Request the week's throughline first — where students should be by Friday — and then have Claude draft Monday's plan in full. Coherence across days is the thing manual planning loses first when you're tired.
• Demand the student-facing materials. The plan is half the job. Ask for the warm-up, the practice set, and the exit ticket as things students actually see, not as a description of them.

Run that once and you have a full first draft in the time it took to describe your class. Which is the point: you spent your energy on the judgment only you have, and handed the transcription to the machine.

The finish line for a lesson is never chat text you have to reformat at 10pm — it's a plan and a set of student materials you can revise, print, and teach. Getting to that artifact fast is the entire value.

Revise, don't accept

A draft is a draft. The workflow assumes you'll push back, and you should. This is where teaching judgment re-enters, and where the tool earns its keep instead of replacing you.

Read the plan the way you'd read a colleague's. Is the warm-up actually a bridge from yesterday, or busywork? Does the worked example use numbers that hide the concept? Is the exit ticket assessing the standard or just the vocabulary? When something's off, say so plainly — "the guided practice jumps too fast, add a scaffolded middle step" — and Claude revises against the same standards and progression it planned from. You're not restarting; you're steering. Because the materials render out cleanly to editable documents, the plan you approve is the plan you print, not a starting point you have to rebuild in a word processor.

Do this across the week and something compounds. By Wednesday, Claude has your corrections in context — your class, your pacing, your non-negotiables. The Thursday draft arrives closer to right. The planning partner learns the room alongside you.

Where it fits in your stack

Planning is one node in a larger loop, and it's worth seeing the whole thing. A week's plan is stronger when it's informed by where students actually are — which is why the analyze-class-data workflow pairs naturally with this one: hand Claude your diagnostics and rosters first, then plan against reality. Once a lesson exists, differentiation is the next move, taking one plan and producing tiered versions for the range of readiness in front of you. And connectors extend the reach — turn a finished plan into polished handouts, or generate aligned practice problems, without leaving the flow.

If you're outside Claude for Teachers but want to experiment, the marketplace has adjacent tooling worth a look — Lesson Plan Studio and the broader education agent skills collection cover the same terrain, and the broader case for why pedagogy lives in a skill rather than a smarter model runs through the AI tutoring skills category.

The habit worth building

The teachers who get the most from this don't treat it as a one-off rescue on a bad night. They build a weekly rhythm. Sit down Sunday, tell Claude what unit you're on and what last week revealed, and generate the week's arc in one session. Revise Monday and Tuesday closely while you learn how it plans; skim Wednesday through Friday once you trust the pattern. Keep the artifacts — the plans accumulate into a bank you can pull from next year, already aligned to your standards.

None of this removes you from planning. Standards alignment, scaffolding, and coherent pacing are still your calls — Claude just makes the first draft real enough that your judgment has something to sharpen instead of a blank page to fill. That's the whole trade the Claude for Teachers series keeps coming back to: give the craft its time back, and spend that time on the students. The lesson-planning workflow is where most teachers feel it first.

Getting verified is free, and the offer runs through June 30, 2027 for a full year of access. If a week of planning is currently costing you a weekend, this is the workflow to start with.

---

Part of the Claude for Teachers series. Related: Standards-Aligned by Construction, Not by Afterthought · Inside k12-lesson-planning: How the Open-Source Skill Works. Browse AI tutoring skills or more builder insights.