# reefapi

One MCP server in front of 160+ ReefAPI web-data engines: search, social, e-commerce, property, jobs, travel, news, finance. Discovery runs keyless; calls need a free key.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · 379 installs
- Author: support-c2w9
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://smithery.ai/server/support-c2w9/reefapi
- Updated: 2026-06-23
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/reefapi

## What it does

reefapi puts a catalogue of 160+ data engines behind a single MCP endpoint. Search, social platforms, shops, property listings, hiring, travel, news and finance sit in one place, and the assistant picks whichever engine fits the task at hand.

Work goes in stages. `search_engines` takes English keywords, or `get_catalog` dumps the whole menu for the model to read itself. Then `get_engine_schema` shows the available actions in compact form, `get_action_schema` spells out every parameter of the one action you chose, and `call_engine` fires the request. Everything comes back in the same envelope: `{ ok, data, meta, error }`.

Schemas stay short on purpose, since an engine with dozens of actions would otherwise flood the context window. One catch: the catalogue is English-only, so a request typed in another language has to be translated before discovery, or the search misses. Lookups of engines and schemas need no credentials. `call_engine` does: a local install reads `REEFAPI_KEY`, while the hosted endpoint at `https://reefapi--support-c2w9.run.tools` expects an `Authorization: Bearer ak_live_...` header.

## Requirements
- Remote server: `https://reefapi--support-c2w9.run.tools`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http reefapi https://reefapi--support-c2w9.run.tools
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add reefapi --url https://reefapi--support-c2w9.run.tools
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "reefapi": {
            "url": "https://reefapi--support-c2w9.run.tools"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "reefapi": {
            "type": "http",
            "url": "https://reefapi--support-c2w9.run.tools"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http reefapi https://reefapi--support-c2w9.run.tools
```

## When to use it
- pull clean JSON from a site that blocks scrapers with captcha or login
- match a task description to the right ReefAPI engine
- get valid call parameters before the first request
- check whether a domain is free without opening a browser
- look up known vulnerabilities in a package

## Example prompts
- Find a ReefAPI engine that returns company reviews and pull the latest ones for me.
- Which engines can detect a site's tech stack? Search the catalogue, then call the right one.
- Run `call_engine` for job listings — the key is already in `REEFAPI_KEY`.

## Similar
- [tavily](https://affpapa.org/en/nevibecode/skills/tavily.md) — skill · 636 installs
- [Scavio](https://affpapa.org/en/nevibecode/mcp/scavio.md) — MCP server · ★ 8 repo stars
- [web-search](https://affpapa.org/en/nevibecode/mcp/magurelabs-web-search.md) — MCP server · ★ 10 repo stars
- [HasData Airbnb](https://affpapa.org/en/nevibecode/mcp/hasdata-airbnb.md) — MCP server · ★ 7 repo stars
- [firecrawl-scrape](https://affpapa.org/en/nevibecode/skills/firecrawl-scrape.md) — skill · 84400 installs
- [firecrawl-knowledge-ingest](https://affpapa.org/en/nevibecode/skills/firecrawl-knowledge-ingest.md) — skill · 31600 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
