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

# n8n

> Send Twexapi MCP handoff JSON into n8n workflows.

Use n8n when you want Twexapi research results to trigger database writes, Slack alerts, CRM updates, or scheduled follow-up jobs.

## Prerequisites

* A Twexapi API key
* n8n Cloud or self-hosted n8n
* A webhook URL for MCP handoff recipes
* Optional Slack, Google Sheets, Airtable, or database credentials

## API key credential

Create a reusable n8n credential for Twexapi REST calls:

| Field          | Value                    |
| -------------- | ------------------------ |
| Authentication | Header Auth              |
| Header name    | `Authorization`          |
| Header value   | `Bearer YOUR_API_KEY`    |
| Base URL       | `https://api.twexapi.io` |

For MCP-capable AI Agent nodes, configure the MCP Client Tool with `https://api.twexapi.io/mcp` and header `x-api-key: YOUR_API_KEY`.

## Pattern

1. Create an n8n Webhook trigger.
2. Ask an MCP-capable agent to return strict Twexapi handoff JSON.
3. POST the JSON to the n8n webhook.
4. Use Set, IF, Split Out, and HTTP Request nodes to process rows.

## Webhook payload

```json theme={null}
{
  "query": "AI agents",
  "route_used": "/twitter/advanced_search",
  "has_more": true,
  "next_cursor": "cursor_123",
  "tweets": [
    {
      "tweet_id": "1803006263529541838",
      "author_username": "openai",
      "created_at": "2026-07-02T10:00:00Z",
      "text": "Example tweet text"
    }
  ]
}
```

## Suggested n8n nodes

| Step | Node               | Purpose                                                               |
| ---- | ------------------ | --------------------------------------------------------------------- |
| 1    | Webhook            | Receive MCP handoff JSON.                                             |
| 2    | IF                 | Stop if `tweets` is empty.                                            |
| 3    | Split Out          | Process each tweet row.                                               |
| 4    | Database or Sheets | Upsert by `tweet_id`.                                                 |
| 5    | HTTP Request       | Continue through Twexapi REST if the workflow owns the next API call. |

## Result handoff

Use an Edit Fields node after each MCP or HTTP Request step when the next node only needs stable fields.

| Source            | Store as          |
| ----------------- | ----------------- |
| Tweet ID          | `tweet_id`        |
| Text              | `text`            |
| Author username   | `author_username` |
| Created time      | `created_at`      |
| Pagination cursor | `next_cursor`     |
| MCP route         | `route_used`      |

## Recipe 1: AI agent X research with MCP

<Steps>
  <Step title="Add an AI Agent node">
    Use your preferred chat model.
  </Step>

  <Step title="Add an MCP Client Tool">
    Configure `https://api.twexapi.io/mcp` with `x-api-key`.
  </Step>

  <Step title="Prompt the agent">
    Ask it to call `explore`, then `twexapi_request`, and return only JSON.
  </Step>

  <Step title="Route rows">
    Split `tweets` and write each row to your destination.
  </Step>
</Steps>

Suggested prompt:

```txt theme={null}
Find recent posts about AI agents in fintech, group them by theme, and list 5 accounts worth monitoring.
Return only JSON with route_used, tweets, has_more, and next_cursor.
```

## Recipe 2: Monitor-style Slack alert

Use this when the workflow should alert a Slack channel after receiving a row from an MCP agent or a scheduled REST job.

| Node            | Purpose                                             |
| --------------- | --------------------------------------------------- |
| Webhook Trigger | Receive Twexapi handoff JSON.                       |
| Split Out       | Iterate over `tweets`.                              |
| Set             | Format `author_username`, `text`, and `public_url`. |
| Slack           | Post the alert.                                     |

Slack message template:

```txt theme={null}
@{{$json.author_username}}
{{$json.text}}
{{$json.public_url || "https://x.com/i/web/status/" + $json.tweet_id}}
```

## Recipe 3: Extraction to Google Sheets

For larger follower or reply exports, let the agent choose the endpoint with MCP, then run the repeated pages through REST.

| Sheet column | Twexapi field                                           |
| ------------ | ------------------------------------------------------- |
| Tweet ID     | `tweet_id`                                              |
| Author       | `author_username`                                       |
| Text         | `text`                                                  |
| Created At   | `created_at`                                            |
| URL          | `public_url` or `https://x.com/i/web/status/{tweet_id}` |

## Direct REST continuation

```txt theme={null}
GET https://api.twexapi.io/twitter/global-trending/tweets
Authorization: Bearer YOUR_API_KEY
```

Keep MCP for agent discovery and handoff. Use REST for scheduled, deterministic n8n jobs.

## Testing checklist

* Use n8n Test Step on every HTTP Request node.
* Confirm every REST request includes `Authorization: Bearer YOUR_API_KEY`.
* Confirm every MCP Client Tool request includes `x-api-key`.
* Confirm webhooks receive JSON, not prose.
* On `429`, pause before retrying and preserve `next_cursor`.