# Spark - AI Assets Marketplace

The Spark - AI Assets Marketplace MCP server searches Spark for prompts, skills, and connectors, then exposes asset details and raw content to your coding agent.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · 828 installs
- Author: entirevc
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://smithery.ai/server/entirevc/spark
- Updated: 2026-03-07
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/spark-ai-assets-marketplace

## What it does

Spark connects an agent to a catalog of prompts, skills, and connectors. Its tools can search the marketplace, fetch an asset record, retrieve the asset payload, list popular entries, and show available categories. Asset records include descriptions, metadata, files, outcome reports, and a link to `spark.entire.vc`.

Searches can be narrowed by domain or popularity. After using an asset, the agent can send one of four outcomes with `report_outcome`: used as received, changed, broken, or not applicable. Paid downloads require authentication, using a Spark API key in the `X-API-Key` header or a Bearer token in the MCP client configuration. The service endpoint is `https://spark.run.tools`.

## Requirements
- Remote server: `https://spark.run.tools`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http spark https://spark.run.tools
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add spark-ai-assets-marketplace --url https://spark.run.tools
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "spark-ai-assets-marketplace": {
            "url": "https://spark.run.tools"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "spark-ai-assets-marketplace": {
            "type": "http",
            "url": "https://spark.run.tools"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http spark-ai-assets-marketplace https://spark.run.tools
```

## When to use it
- search the marketplace by domain
- inspect an asset before using it
- retrieve raw prompt or skill content
- find popular assets and available categories
- record what happened after applying an asset

## Example prompts
- Search Spark for prompt assets that help with PDF workflows using `search_assets`.
- Use `get_asset` to inspect the asset with slug `customer-support`, then retrieve its raw content with `get_asset_content`.
- After trying the asset, report that it broke by calling `report_outcome`.

## Similar
- [finding-data-lake-assets](https://affpapa.org/en/nevibecode/skills/finding-data-lake-assets.md) — skill · 7400 installs
- [ue-data-assets-tables](https://affpapa.org/en/nevibecode/skills/ue-data-assets-tables.md) — skill · 1500 installs
- [spark-optimization](https://affpapa.org/en/nevibecode/skills/spark-optimization.md) — skill · 10700 installs
- [google-cloud-solution-agentic-analytics-spark-knowledge-catalog](https://affpapa.org/en/nevibecode/skills/google-cloud-solution-agentic-analytics-spark-knowledge-catalog.md) — skill · 5000 installs
- [spark-engineer](https://affpapa.org/en/nevibecode/skills/spark-engineer.md) — skill · 4200 installs
- [marketplace-search-recsys-planning](https://affpapa.org/en/nevibecode/skills/marketplace-search-recsys-planning.md) — skill · 211 installs

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
