Free tools for teachers

Four tools, built around what AI can't do for a child

Each tool starts from a capacity children build through effort and cannot build any other way: reaching a hard objective by a route that fits them, chasing a question that is genuinely theirs, seeing across the boundaries between subjects, and doing something real with what they know.

Below is why each of those is important and what the tool actually does to support it.

Universal Design for Learning

Differentiation often lowers the bar

The instinct, when a student cannot reach an objective the way a lesson is built, is to give them a smaller objective. Shorter text, fewer problems, an easier version of the task. The student who receives it is being told, in a way nobody says out loud, that the hard thinking is for other students.

Universal Design for Learning changes this approach. The barrier is what changes, never the demand. A student who cannot carry the reading load still has to construct the argument. A student who cannot write at length still has to justify the answer. The route varies but the destination does not.

The reason this is hard in practice is arithmetic. Designing three genuine ways into an objective, three ways through it, and three ways to demonstrate it, for one lesson, is an hour of work a teacher with five preps does not have. So differentiation collapses into the easier version, not because anyone believes in it, but because it is the only version that fits in the constraints.

CAST's UDL Guidelines organise this across engagement, representation, and expression. Ruesch and Sarvary (2024) found that structure and flexibility held together, rather than traded off, is what makes a learning environment genuinely inclusive.
UDL Architect

Drafts nine options against one objective you name as non-negotiable: three ways in, three ways through, three ways to show it. You choose one door per column, and it writes the full menu around your choices. Every option is checked against the objective it serves, and each one states how it holds the cognitive bar while lowering its barrier.

Open UDL Architect →

Self-directed learning

The question a child actually asked

Children arrive with questions that have nothing to do with the unit. Why do cats purr. How does code work. What did Korean kings believe. In many classrooms these are interruptions, acknowledged warmly and then set aside, because the curriculum is already written and the day is already full.

But children need the practice of sustaining an investigation they chose, through the part where it stops being fun, toward something they can show another person. That is a capacity, and it is built by doing it. A child who has only ever pursued questions assigned to them does not develop intrinsic curiosity, and by the time anyone notices, they are the student who says I don't know what to do when the instructions run out.

I've had students build working airplanes, dye maps with tea to age them, teach themselves to code, and make gymnastics tutorials in Canva. None of it came from a lesson plan.

Gruber and colleagues (2014) found that curiosity changes what the brain retains, with states of curiosity improving memory through the dopaminergic circuit. Mitra's self-organised learning environments (2006; Dolan et al., 2013) showed children sustaining genuine investigation with boundaries and encouragement rather than instruction, and Kruger, Werlen and Bergamin (2022) tie that persistence to emotional connection with the question itself.
Wonder Web

Takes a real question a student asked and maps it into six sub-questions the student can actually investigate. You choose which threads they chase, and it writes a two-week pathway plan: what the student does, where you check reasoning rather than conclusions, and how it ends on a question still open. Nothing in it answers the question for them.

Open Wonder Web →

Cross-domain synthesis

What math class leaves in math class

Students learn to represent data in a graph, and then cannot reach for that when they collect survey responses for a persuasive essay or need to show an experimental result. The capacity to transfer exists. Subjects are organised in silos so cross-domain synthesis rarely gets attempted.

Compartmentalisation is a rational response to how school is built. When each subject is graded separately, when no task ever requires two of them at once, a student correctly concludes that connection-making does not count. The failure is not a bad interdisciplinary project. It is a student who cannot see a connection that is obvious to everyone else.

Most attempts to fix this produce decoration. Word problems about music are not mathematics meeting music; they are mathematics wearing a costume. A genuine bridge is a concept both subjects are actually about, and the test is simple: remove either subject, and the lesson has to collapse. Ratio is a real bridge between fraction work and harmony, because halving a string length is doubling a pitch and there is no way to teach one half of that.

Fauconnier and Turner (2002) describe this as conceptual blending, the combination of elements from separate domains into something neither contained. Thomas (2024) found measurable gains in fluency, flexibility, originality and elaboration when K-6 students practised divergent thinking across subjects, and Liu and colleagues (2022) found interdisciplinary teaching improved creativity in a controlled comparison.
CrossLink

Drafts three genuine conceptual bridges between two subjects you name, each with the reason it holds and the angle each subject contributes. You pick the one that survives the collapse test, and it builds the full interdisciplinary lesson on it, with both subjects doing real work rather than one decorating the other.

Open CrossLink →

Community connection

The standard that never leaves the worksheet

Every teacher has been asked when a student will ever use this, and most of us have answered, for the next grade. We assess standards based on the curriculum and the worksheets. Students can tell when they feel no meaning, which is why the question keeps coming back.

A standard is genuinely applied when it is at work somewhere in the world the student already knows, and someone in that world would recognise what the class is doing. Produce sellers compare fractions of a kilo against prices all day. A persuasive letter is a real form when it goes to someone who can actually say yes: the principal who controls the schedule, the shop on the corner, the municipality that decides where the bins go. The standard is the same standard. What changes is that somebody answers.

The barrier here is that finding the real application, and then finding the person who does it for a living, takes local knowledge and time that a teacher covering a full curriculum does not have.

This is where the tool is most clearly a draft rather than an answer. It proposes applications; you know which ones your students would actually recognise from their own lives, and no model does.
REAL Connections

Drafts five authentic applications of a standard in the world your students live in, and three types of community partner who work with it daily. You choose which are real for your room, and it writes the activity set plus a draft outreach note you can adapt and send yourself. The real-world setting raises the stakes; it does not replace the thinking.

Open REAL Connections →

One standard across all four

Each tool drafts options and stops. You compose from them, and only then does it write the full artifact from what you chose. That two-stage shape is deliberate: a tool that hands over a finished lesson has made the professional decisions on your behalf, and the decisions are the job.

Nothing in the suite is built to reach a student directly. Anything a teacher might say to a class arrives labelled as draft wording, for you to adapt in your own voice. And no tool answers for a child or smooths away the struggle a lesson exists to produce, because that struggle is where the capacity gets built.

Free, and built for how you already work. No account, no login, no student data. The tools ask for your objective or your standard, never for a child's name or record. Open one, work through it, and keep what you make.

Start from the job you already have.
Open the Intelligence Suite →