> ## Documentation Index
> Fetch the complete documentation index at: https://docs.learningcommons.org/llms.txt
> Use this file to discover all available pages before exploring further.

# K-12 Lesson Prep

> Reference for the K-12 Lesson Prep skill, which helps teachers internalize an existing lesson through guided analysis of sample student work.

[Skill last updated August 28, 2026](https://github.com/learning-commons-org/agent-skills/tree/main/skills/k12-lesson-prep)

<Note>
  The K-12 Lesson Prep Agent Skill was co-developed with
  [Anthropic](https://www.anthropic.com/) ↗.
</Note>

## Overview

The [K-12 Lesson Prep](https://github.com/learning-commons-org/agent-skills/tree/main/skills/k12-lesson-prep) ↗ skill helps a teacher prepare for, and internalize, an existing lesson before teaching it:

* Build mastery of what's important about the lesson and why
* Anticipate how to distinguish between student mastery and a need for more support
* Walk into class confident in their understanding of the lesson's content and focus

This skill's output is not a rendered file – it is the conversation itself. The skill presents the teacher one small piece of student work at a time, then waits for the teacher's read before offering its own. A short teacher-only prep note may be produced at the end, but this is a byproduct of the conversation, not the point of it.

<Note>
  This skill works with or without Knowledge Graph, but we strongly recommend
  connecting to it. See [Quickstart: Connect Knowledge
  Graph](/agent-skills/getting-started/quickstart#connect-knowledge-graph) for
  more information.
</Note>

## At a glance

|                      |                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                       |
| :------------------- | :------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
| **Workflow**         | Internalize an existing lesson through conversation before teaching it                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                |
| **Subjects**         | Math, ELA, Science, Social Studies                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                    |
| **Supported grades** | K–12                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                  |
| **Rubrics**          | <ul><li>[`shared.csv`](https://github.com/learning-commons-org/agent-skills/blob/main/evals/k12-lesson-prep/rubrics/shared.csv) ↗ — Core criteria: teacher-first sequencing, near-miss anchoring, no unsolicited student-level inference, one-glance note structure</li><li>[`confirmation_bias.csv`](https://github.com/learning-commons-org/agent-skills/blob/main/evals/k12-lesson-prep/rubrics/confirmation_bias.csv) ↗ — Scorer for whether the model avoids extending or confirming theories about named students</li><li>[`student_work_first.csv`](https://github.com/learning-commons-org/agent-skills/blob/main/evals/k12-lesson-prep/rubrics/student_work_first.csv) ↗ — Scorer for whether the teacher is asked to respond before the model shares its own read</li></ul> |

## Getting started

Follow the [Quickstart](/agent-skills/getting-started/quickstart) to start using this skill:

| Access method           |                                                                                                                              |
| :---------------------- | :--------------------------------------------------------------------------------------------------------------------------- |
| **Knowledge Graph MCP** | [Quickstart: Connect Knowledge Graph](/agent-skills/getting-started/quickstart#connect-knowledge-graph)                      |
| **Example prompts**     | [example-prompts.md](https://github.com/learning-commons-org/agent-skills/blob/main/skills/example-prompts.md#lesson-prep) ↗ |
| **Skill instructions**  | [`SKILL.md`](https://github.com/learning-commons-org/agent-skills/blob/main/skills/k12-lesson-prep/SKILL.md) ↗               |

## Triggers

Your LLM will programmatically load this skill when explicitly or implicitly prompted to prepare to teach a specific lesson that already exists:

* **Explicitly**: "prep me for tomorrow's lesson," "get me ready to teach this"
* **Implicitly**: "I'm teaching this lesson tomorrow and want to feel ready," "what should I be watching for when I teach this," "help me get ready for class"

## Output

| Artifact                            | Description                                                                                                                                                                                 |
| :---------------------------------- | :------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
| **Conversation itself**             | Primary output: a turn-by-turn walkthrough of the lesson's near-miss vs. proficient student work, its highest-leverage sticking points, and how to respond without de-rigoring the material |
| **Optional teacher-only prep note** | A short, scannable note in the lesson's own teaching order, produced only on request or at the natural end of the conversation                                                              |

## Scoring

1. Run a lesson prep conversation using this skill.
2. Apply `shared.csv` plus `confirmation_bias.csv` and `student_work_first.csv`, which evaluate conversational behavior rather than a finished artifact.
3. Run LLM-as-judge or human review against each criterion. See [Evaluating outputs](/agent-skills/using-agent-skills/evaluating-outputs).

## Skill release history

| Date            | Changed                                    |
| --------------- | ------------------------------------------ |
| August 28, 2026 | First release, co-developed with Anthropic |