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

# Agent Skills

> Connect Pinecone and Agent Skills to ship vector search and RAG: embed, index, and query at scale with managed infrastructure.

export const PrimarySecondaryCTA = ({primaryLabel, primaryHref, primaryTarget, secondaryLabel, secondaryHref, secondaryTarget}) => <div style={{
  display: 'flex',
  alignItems: 'center',
  gap: 16
}}>
   {primaryLabel && primaryHref && <div style={{
  width: 'fit-content',
  height: 42,
  background: 'var(--brand-blue)',
  borderRadius: 4,
  overflow: 'hidden',
  flexDirection: 'column',
  justifyContent: 'center',
  alignItems: 'center',
  display: 'inline-flex'
}}>
      <a href={primaryHref} target={primaryTarget} style={{
  paddingLeft: 22,
  paddingRight: 22,
  paddingTop: 8,
  paddingBottom: 8,
  justifyContent: 'center',
  alignItems: 'center',
  gap: 4,
  display: 'inline-flex',
  textDecoration: 'none',
  borderBottom: 'none'
}}>
        <div style={{
  textAlign: 'justify',
  color: 'var(--text-contrast)',
  fontSize: 15,
  fontWeight: '600',
  letterSpacing: 0.46,
  wordWrap: 'break-word'
}}>
          {primaryLabel}
        </div>
        <svg width="24" height="24" viewBox="0 0 24 24" fill="none" xmlns="http://www.w3.org/2000/svg" style={{
  marginLeft: 2
}}>
          <path d="M9.70492 6L8.29492 7.41L12.8749 12L8.29492 16.59L9.70492 18L15.7049 12L9.70492 6Z" fill="white" style={{
  fille: "var(--text-contrast)"
}} />
        </svg>
      </a>
    </div>}

    {secondaryLabel && secondaryHref && <div style={{
  width: 'fit-content',
  height: 42,
  borderRadius: 4,
  overflow: 'hidden',
  flexDirection: 'column',
  justifyContent: 'center',
  alignItems: 'center',
  display: 'inline-flex',
  textDecoration: 'none'
}}>
        <a href={secondaryHref} target={secondaryTarget} style={{
  paddingLeft: 11,
  paddingRight: 11,
  paddingTop: 8,
  paddingBottom: 8,
  justifyContent: 'center',
  alignItems: 'center',
  gap: 8,
  display: 'inline-flex',
  textDecoration: 'none',
  borderBottom: 'none'
}}>
          <div style={{
  textAlign: 'justify',
  color: 'var(--brand-blue)',
  fontSize: 15,
  fontWeight: '600',
  letterSpacing: 0.46,
  wordWrap: 'break-word'
}}>
            {secondaryLabel}
          </div>
        </a>
      </div>}

  </div>;

Pinecone's official [Agent Skills](https://github.com/pinecone-io/skills) library brings Pinecone capabilities to any agentic IDE that supports the Agent Skills standard. Use skills to manage indexes, run semantic search, create document Q\&A assistants, and more — all through natural language in your IDE.

Compatible with [Cursor](https://www.cursor.com/), [GitHub Copilot](https://github.com/features/copilot), [Codex](https://chatgpt.com/codex), [Gemini CLI](https://github.com/google-gemini/gemini-cli), and other agentic IDEs.

<PrimarySecondaryCTA primaryLabel={"View on GitHub"} primaryHref={"https://github.com/pinecone-io/skills"} primaryTarget={"_blank"} />

<Tip>
  If you use **Claude Code**, install the dedicated [Pinecone plugin for Claude Code](/integrations/claude-code) instead. If you use **Gemini CLI**, install the dedicated [Pinecone extension for Gemini CLI](/integrations/gemini-cli) instead. Both include additional features specific to those tools.
</Tip>

## Features

* **7 built-in skills** for index management, semantic search, assistant creation, and more
* **Universal compatibility** with any IDE that supports Agent Skills
* **Works with the Pinecone MCP server** for direct index operations

## Prerequisites

* A [Pinecone API key](https://app.pinecone.io/organizations/-/keys)
* [Node.js](https://nodejs.org/) installed (for `npx`)
* [Pinecone MCP server](/guides/operations/mcp-server) configured in your IDE (optional, enables the `query` skill)
* [uv](https://docs.astral.sh/uv/getting-started/installation/) installed (optional, runs bundled Python scripts)
* [Pinecone CLI](/reference/cli/quickstart) installed (optional, enables the `cli` skill)

## Installation

<Steps>
  <Step title="Set your API key">
    ```shell theme={null}
    export PINECONE_API_KEY="YOUR_API_KEY"
    ```

    Replace `YOUR_API_KEY` with your [Pinecone API key](https://app.pinecone.io/organizations/-/keys).
  </Step>

  <Step title="Install the skills">
    ```shell theme={null}
    npx skills add pinecone-io/skills
    ```

    This downloads Pinecone's skills into your project, making them available to your IDE's AI agent.
  </Step>

  <Step title="Configure the MCP server (optional)">
    For full functionality, configure the [Pinecone MCP server](/guides/operations/mcp-server) in your IDE. This enables the `query` skill and direct index operations.
  </Step>
</Steps>

## Available skills

| Skill             | Description                                                                                 |
| ----------------- | ------------------------------------------------------------------------------------------- |
| **quickstart**    | Step-by-step onboarding — create an index, upload data, and run your first search.          |
| **query**         | Search integrated indexes using natural language text via the Pinecone MCP.                 |
| **assistant**     | Create, manage, and chat with Pinecone Assistants for document Q\&A with citations.         |
| **cli**           | Use the Pinecone CLI for terminal-based index and vector management across all index types. |
| **mcp**           | Reference for all available Pinecone MCP server tools and their parameters.                 |
| **pinecone-docs** | Curated links to official Pinecone documentation, organized by topic.                       |
| **help**          | Overview of all skills and what you need to get started.                                    |

## MCP tools

Agent Skills work alongside the [Pinecone MCP server](/guides/operations/mcp-server), which provides tools for listing indexes, creating indexes, upserting records, searching, reranking, and more. Configure the MCP server in your IDE to enable the `query` skill and direct index operations.

For the full list of MCP tools, see [Use the Pinecone MCP server](/guides/operations/mcp-server).

## Resources

* [GitHub repository](https://github.com/pinecone-io/skills)
* [Pinecone MCP server guide](/guides/operations/mcp-server)