> ## 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.

# Quickstart

> Get started with Agent Skills: install skills, connect Knowledge Graph, and try example prompts.

## What you'll do

* [Install Agent Skills](#install-skills) to your agent runtime
* [Connect Knowledge Graph](#connect-knowledge-graph) for the most accurate outputs (Recommended)
* Generate a standards-aligned K-12 lesson
* [Evaluate your output](/agent-skills/getting-started/evaluating-outputs) to benchmark it against published rubrics

## What you'll need

* An agent runtime that supports the [open Agent Skills format](https://agentskills.io/) ↗
  * Claude Code
  * Cursor
  * Codex
* A [Learning Commons Platform](https://platform.learningcommons.org/?utm_source=docs\&utm_medium=agent-skills) ↗ API key <Badge>For Knowledge Graph integration</Badge>

## Steps

<Steps>
  <Step title="Install skills">
    <span id="install-skills" />

    <Note>
      Installation behavior may vary slightly by runtime. Check your agent's
      documentation for how it discovers and loads skills from installed packages.

      **For Claude Code**: if your project doesn't have a `.claude/` directory yet, run the following command first before proceeding with installation:

      ```bash theme={null}
      mkdir -p .claude
      ```
    </Note>

    Install [`agent-skills`](https://github.com/learning-commons-org/agent-skills?utm_source=docs\&utm_medium=agent-skills) ↗ into your agent runtime of choice:

    ```shell theme={null}
    npx skills add learning-commons-org/agent-skills
    ```

    | Path                                                                                                                                                                              | Description                             |
    | :-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | :-------------------------------------- |
    | [skills/k12-lesson-planning/](https://github.com/learning-commons-org/agent-skills/tree/main/skills/k12-lesson-planning?utm_source=docs\&utm_medium=agent-skills) ↗               | Standards-aligned lesson planning skill |
    | [skills/k12-lesson-differentiation/](https://github.com/learning-commons-org/agent-skills/tree/main/skills/k12-lesson-differentiation?utm_source=docs\&utm_medium=agent-skills) ↗ | Tiered lesson differentiation skill     |
    | [skills/example-prompts.md](https://github.com/learning-commons-org/agent-skills/blob/main/skills/example-prompts.md?utm_source=docs\&utm_medium=agent-skills) ↗                  | Example prompts by subject and workflow |

    Each skill includes:

    * `SKILL.md` — Metadata and step-by-step instructions
    * `references/` — Subject-specific pedagogy and output mappings
    * `scripts/` — Render helpers for `.docx` Word documents
  </Step>

  <Step title="Connect Knowledge Graph">
    <span id="connect-knowledge-graph" />

    <Note>
      Knowledge Graph datasets carry varying licenses. Some datasets aren’t readily accessible by default due to license requirements.
    </Note>

    Add the [Knowledge Graph MCP server](/knowledge-graph/using-knowledge-graph/mcp-server) to your agent runtime (replace `$LC_API_KEY` with your Learning Commons Platform API key):

    ```shell Example for Claude Code theme={null}
    claude mcp add --transport http learning-commons-kg \
      https://kg.mcp.learningcommons.org/mcp \
      --header "Authorization: Bearer $LC_API_KEY"
    ```

    Access to [Knowledge
    Graph](/knowledge-graph/understanding-knowledge-graph/introduction) data lets
    the agent:

    * Resolve [Academic Standards](/knowledge-graph/graph-reference/academic-standards)
    * Understand granular [Learning Components](/knowledge-graph/graph-reference/learning-components)
    * Trace [Learning Progressions](/knowledge-graph/graph-reference/learning-progressions)
    * Find aligned [Curriculum](/knowledge-graph/graph-reference/curriculum) lessons and common misconceptions

    <Note>
      When Knowledge Graph is unavailable, the skills still run and fall back to the
      model's general knowledge. However, outputs grounded in Knowledge Graph are
      more accurate and better aligned to specific state standards.
    </Note>
  </Step>

  <Step title="Verify the connector">
    Start a new agent session and confirm that Knowledge Graph tools (e.g., `find_standard_statement`) are available.

    If the connector is not available, Agent Skills still produce usable output without Knowledge Graph data. They will not invent Knowledge Graph citations or attribute content to curriculum materials they have not retrieved.

    <Warning>
      When Knowledge Graph is connected, skills must call MCP tools before drafting.
      Not calling when connected is treated as a critical failure in the skill
      instructions.
    </Warning>
  </Step>

  <Step title="Prompt your agent">
    <span id="prompt-your-agent" />

    Make a typical teaching request. Some [example prompts](https://github.com/learning-commons-org/agent-skills/blob/main/skills/example-prompts.md?utm_source=docs\&utm_medium=agent-skills) ↗ include:

    * *"I need a lesson for tomorrow on rounding to the nearest hundred for my 3rd graders."*
    * *"Differentiate this 6th grade food webs lesson for students below / at / and above proficiency level (find the lesson here: [https://www.calacademy.org/educators/lesson-plans/how-stable-is-your-food-web](https://www.calacademy.org/educators/lesson-plans/how-stable-is-your-food-web))"*

    Your agent should load the matching Agent Skill automatically.
  </Step>
</Steps>