Skip to main content
Skill last updated July 14, 2026
The K-12 Lesson Planning Agent Skill was co-developed with Anthropic ↗.

Overview

The K-12 Lesson Planning skill creates a lesson plan, student-facing materials, and an observation template for math, ELA, science, or social studies. It produces teacher-ready, standards-aligned artifacts as native files in Turn 1, rendered from one master JSON via bundled scripts (editable .docx Word documents on confirmation). The skill works with or without the Learning Commons Knowledge Graph. The skill considers:
  • Subject routing (math, ELA, science, social studies) with mandatory subject reference files
  • Curriculum alignment when a teacher names or implies a curriculum (for example, Illustrative Mathematics or OpenSciEd)
  • Standards grounding via Knowledge Graph when connected
  • Subject-specific pedagogy, grade-band structures, and output mappings
Do not load this skill when the request is only for grading, a rubric, assessment feedback, a quiz, or a standards lookup — answer those directly without it.

At a glance

Getting started

Follow the Quickstart to start using this skill:

Triggers

Your LLM will programmatically load this skill when explicitly or implicitly prompted to build a K-12 lesson from scratch:
  • Explicitly: “math lesson plan,” “mini-lesson,” “unit plan,” “reading lesson,” “phonics lesson”
  • Implicitly: “I’m teaching long division to 5th graders,” “planning a lesson on RL.4.3 next week,” “need to teach photosynthesis tomorrow”
Core signal: teacher has a topic/standard + grade and needs new instructional content created.

Output

Rubrics

The K-12 Lesson Planning skill’s outputs are scored by the following rubrics ↗.
To generate lesson plans, apply the shared.csv rubric first, then layer in the relevant subject-specific file to extend the criteria set.Example: For a 7th grade ELA lesson, you would score against shared.csv and ela.csv.

How to score

  1. Generate lesson materials using this skill.
  2. Select shared.csv plus the subject-specific rubric that matches the lesson.
  3. Run LLM-as-judge or human review against each criterion. See Evaluating outputs.
See ATTRIBUTION.md ↗ in the repository for source credits.

Skill release history