### Install Qdrant Skills CLI

Source: https://github.com/qdrant/skills/blob/main/README.md

Install the Qdrant Skills using the npx skills CLI. This is a common method for adding agent skills.

```bash
npx skills add qdrant/skills
```

--------------------------------

### Example Documentation Link

Source: https://github.com/qdrant/skills/blob/main/AGENTS.md

Demonstrates how to link to Qdrant documentation within skill descriptions. This format is used for inline references to specific features or concepts.

```markdown
- Enable scalar quantization with `always_ram=true` [Scalar quantization](https://search.qdrant.tech/md/documentation/manage-data/quantization/?s=scalar-quantization)
```

--------------------------------

### Install Qdrant Skills for Claude Code

Source: https://github.com/qdrant/skills/blob/main/README.md

Add the marketplace and install all Qdrant skills for Claude Code. This method involves using specific commands within the Claude Code environment.

```bash
/plugin marketplace add qdrant/skills
/plugin install qdrant@qdrant
```

--------------------------------

### Example of a Bad Skill Structure (Integration Guide)

Source: https://github.com/qdrant/skills/blob/main/CONTRIBUTING.md

Presents a skill that acts as an integration guide, which is also not the intended purpose. Skills should focus on diagnostic guidance and operational judgment.

```markdown
## Integrating Qdrant with Framework X
- Install the framework package
- Configure the vector store
- Run a similarity search
```

--------------------------------

### Example of a Bad Skill Structure (Tutorial)

Source: https://github.com/qdrant/skills/blob/main/CONTRIBUTING.md

Shows a skill that resembles a tutorial, which is not the intended purpose of Qdrant Skills. Skills should provide judgment, not step-by-step instructions.

```markdown
## Multimodal RAG: Building Document Search
- Build a RAG system using embeddings and Ollama for generation
- Implement basic retrieval from a collection
```

--------------------------------

### Example of a Good Skill Structure

Source: https://github.com/qdrant/skills/blob/main/CONTRIBUTING.md

Illustrates a well-formed skill that addresses a user's problem with actionable steps and links to relevant documentation.

```markdown
## What to do if memory usage is too high?
- Check collection parameters [link to docs]
- Apply quantization [link to choosing quantization]
- Monitor memory usage in prod [link to grafana dashboard]
```

--------------------------------

### Skill File Structure

Source: https://github.com/qdrant/skills/blob/main/CONTRIBUTING.md

Defines the standard directory and file structure for organizing Qdrant Skills and their sub-skills.

```bash
skills/
  <skill-name>/
    SKILL.md              # skill definition (frontmatter + guidance)
    <sub-skill>/
      SKILL.md            # sub-skill for a specific topic
```

--------------------------------

### Search for Qdrant Code Snippets

Source: https://github.com/qdrant/skills/blob/main/skills/qdrant-clients-sdk/SKILL.md

Use this curl command to search for code snippets for a specific language and query. Available languages include python, typescript, rust, java, go, and csharp.

```bash
curl -X GET "https://snippets.qdrant.tech/search?language=python&query=how+to+upload+points"
```

--------------------------------

### Upload Points using Python qdrant-client

Source: https://github.com/qdrant/skills/blob/main/skills/qdrant-clients-sdk/SKILL.md

Uploads multiple vector-embedded points to a Qdrant collection using the Python qdrant_client. Supports parallel uploads and retries. The operation is idempotent.

```python
client.upload_points(
    collection_name="{collection_name}",
    points=[
        models.PointStruct(
            id=1,
            payload={
                "color": "red",
            },
            vector=[0.9, 0.1, 0.1],
        ),
        models.PointStruct(
            id=2,
            payload={
                "color": "green",
            },
            vector=[0.1, 0.9, 0.1],
        ),
    ],
    parallel=4,
    max_retries=3,
)
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.