> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nativeport.ai/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nativeport.ai/_mcp/server.

# Parallel

Fronts the Search API of [Parallel](https://parallel.ai); the gateway supplies Parallel's `/v1` base itself:

```
POST /parallel/search   → https://api.parallel.ai/v1/search
```

Only `POST` is mounted. Controls such as `objective`, `search_queries`, and `processor` live in the JSON body, which is relayed without modification.

Task, Extract, Chat, Monitor, and Find All — the rest of Parallel's surface — stay **intentionally unexposed**; nothing but the synchronous Search API gets proxied.

## Credential handling

* **`x-api-key` is injected**; a client-supplied value is overwritten.
* **`Authorization` is dropped** before the request goes upstream — your gateway token is meaningless to Parallel.

## Example

```bash
curl -X POST "$GATEWAY/parallel/search" \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"objective": "Find recent news about Cloudflare Workers", "search_queries": ["cloudflare workers news"]}'
```

Responses arrive as Parallel's JSON, untouched. Field-level request and response detail lives in [Parallel's documentation](https://docs.parallel.ai); the gateway modifies neither.