Insights

Most AI lesson tools write first and align later. Claude for Teachers draws standards and learning progressions at planning time, so lessons come out aligned and scaffolded by construction — not retrofitted with a standard code stapled on at the end.

Both open-source Claude for Teachers skills ship per-subject reference files — ela.md, math.md, science.md, social_studies.md, plus learning-commons-kg.md. Here's why subject-specific guidance beats one generic prompt.

You control what class data goes into Claude, and nothing you share trains the model. Here's a practical, teacher-tested checklist for handling rosters, diagnostics, and notes responsibly.

TeachFX's Claude for Teachers connector gives personalized instructional feedback grounded in real classroom talk — feedback on your teaching, from your actual room, not a once-a-year observation.

Scheduled tasks in Claude for Teachers let you hand off reviewing each day's exit tickets and adapting tomorrow's plan — running every school day at 4pm, so the analysis is done by the time you get home.

Anthropic's train-the-trainer module, co-created with the American Federation of Teachers, solves a harder problem than the individual course: how one prepared person brings a whole staff along.

Stop asking whether AI is good for the classroom. Ask which tasks it suits. A practical, model-agnostic framework from Anthropic's CC-licensed teacher guidance — and the three-question gut check behind it.

The single hardest thing to make an AI tutor do is not answer. Withholding the solution and asking the next guiding question — the Socratic move — is the core design constraint that separates a tutor from an answer key.

A smarter base model doesn't automatically make a better tutor. Pedagogy is a behavior you install — a SKILL.md that tells the agent when to question, when to scaffold, when to assess — not something you get from more parameters.

Point AI at assessment instead of answers and it becomes a formative-assessment engine — generating quizzes, blueprints, and retrieval practice that reveal what a learner doesn't yet know.

You don't need forty education skills. A working AI tutor is four composable behaviors — question, explain-and-ground, drill, and assess — assembled from the marketplace. Here's the stack.

LLMs are extraordinary analogy engines — the core of a good explainer. They're also confidently wrong in ways a novice can't catch. Great AI explanation is analogy plus grounding, never analogy alone.