# How do I find the posts on a topic that took off on X (Twitter) with AI?

Connect SocialToAI to your AI client over MCP, then ask in plain words. In this recorded run an AI client answered “Which posts about AI agents took off on X this week?” with 1 request on X: 31 items read in 14.4 seconds for 24 credits ($0.08). Below are the calls it made, the answer with its sources, and the prompt and API calls to run it yourself. Recorded 2026-10-18.

- Page: https://socialtoai.com/docs/guides/find-viral-posts-on-x/
- Last updated: 2026-10-18

## The question

> Which posts about AI agents took off on X this week?

X ranks posts by what it thinks you want, and a post's numbers mean little until they are set against the other posts on the same topic. Reading what this run returned by hand takes about 25 minutes (5,905 words at 240 a minute).

## What the AI did

It chose the tools itself: 1 request on X. Each row is one tool call as the client showed it, with what it sent, how much came back and what it cost.

| Platform | Tool | Sent | Requests | Returned | Credits |
| --- | --- | --- | --- | --- | --- |
| X | [`research viral_finder`](https://socialtoai.com/tools/viral-finder.md) | `query: "AI agents" · window: "7d" · scale: "small"` | 1 | 31 posts | 24 |
| **Total** |  |  | 1 | 31 items | **24** |

## The answer

What the client answered, from what it had read. The line under each finding is a post it read, or a figure from those posts.

Recorded 2026-10-18

31 posts found; 7 did at least 3 times the usual. The formats that took off most, each with its share of the 31 posts the analysis could name:

1. **Showcase** (23%, 7 posts): [“transformed an ordinary academic reading list into an interactive research lab right inside the browser”](https://x.com/Bober_smart/status/2110080979742192101)
2. **Story** (10%, 3 posts): [“Grok Bot built me a completely autonomous business which has already generated $1,100”](https://x.com/AlexFinn/status/2110181971280433613)
3. **List** (29%, 9 posts): [“10 GitHub skills + tools that make Hermes instantly more useful.”](https://x.com/BentoBoiNFT/status/2110111219172712922)

Used 24 credits · $0.08. Every claim links to its source.

### The posts it rests on

A selection of what was read; the marked words are what a finding rests on. An item without a link says so.

- X · post · 4d: “t is becoming f*cking dangerous: AI agents from Claude Code transformed an ordinary academic reading list into an interactive research lab right inside the browser Boris Cherny, who created Claude Code, gave a doz” ([Open the source](https://x.com/Bober_smart/status/2110080979742192101))
- X · post · 4d: “Grok Bot built me a completely autonomous business which has already generated $1,100 I think this is the future of the economy Everyon” ([Open the source](https://x.com/AlexFinn/status/2110181971280433613))
- X · post · 4d: “10 GitHub skills + tools that make Hermes instantly more useful. Pick the tools for each bottleneck: research, wri” ([Open the source](https://x.com/BentoBoiNFT/status/2110111219172712922))

## What it cost

1 request, 24 credits in total: $0.08 at the $10 rate. Every answer of the API states the credits it charged and the balance left, so your AI can tell you what a question cost. A call that fails costs nothing.

| Pack | Credits per dollar | This answer |
| --- | --- | --- |
| $10 = 3,000 credits | 300 | $0.08 |
| $50 = 20,000 credits | 400 | $0.06 |
| $200 = 100,000 credits | 500 | $0.048 |

## Ask it yourself

Connect SocialToAI to your AI client once ([the steps for each client](https://socialtoai.com/mcp.md)), then paste the question. You do not name tools or parameters: the client picks them.

**Prompt**

```
Which posts about AI agents took off on X this week?
```

The same in code: the calls this run made, as requests to the API.

**tool.viral_finder**

```
curl https://api.socialtoai.com/v1/tools/viral_finder \
  -H "Authorization: Bearer $SOCIALTOAI_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "query": "AI agents",
    "platform": "x",
    "window": "7d",
    "scale": "small"
  }'
```

## What to try next

- Ask next: “Which hooks do the viral posts open with?”
- Ask next: “Check again next week”
- Ask what the viral posts' first lines have in common.
- Run it again next week.
- Run it as one request with the ready-made tool [Viral finder](https://socialtoai.com/tools/viral-finder.md): Posts on a subject that did clearly better than their authors usually do, on one platform.

- [X (Twitter) MCP server](https://socialtoai.com/mcp/x.md): What your AI can read on X, and how to connect it.
- [X (Twitter) API](https://socialtoai.com/platforms/x.md): Every call on X, with its credits and a request you can copy.