> ## Documentation Index
> Fetch the complete documentation index at: https://developers.t2000.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Deep-dive a crypto asset

> One token, the full picture. Market data + headlines + a web-grounded narrative with citations — CoinGecko + NewsAPI + Perplexity, paid as you go.

Beyond a price check: your agent pulls market data, recent news, and a citation-backed narrative, then synthesizes a single-asset deep-dive. Useful when you want the *why* behind the number, not just the number. \~\$0.08.

***

## The prompt

```
Use t2 services. Deep-dive SUI for me — market data from CoinGecko, recent headlines from
NewsAPI, and a web-grounded narrative with citations from Perplexity. Then a
200-word take.
```

***

## What runs

1. `POST /coingecko/v1/markets` — price, market cap, volume, 7d sparkline (\~\$0.02)
2. `POST /newsapi/v1/search` — recent headlines (\~\$0.02)
3. `POST /perplexity/v1/chat/completions` — web-grounded narrative with live citations (\~\$0.02)
4. `POST /anthropic/v1/messages` — Claude fuses it into the final take (\~\$0.02)

***

## Run it

### Claude Desktop (MCP)

```bash theme={null}
npm install -g @t2000/cli && t2 init && t2 fund && t2 mcp install
```

Paste the prompt with any asset (*"Deep-dive ETH"*, *"What's the bull/bear case for SUI?"*).

### SDK

```typescript theme={null}
import { T2000 } from '@t2000/sdk';

const agent = await T2000.create();
const coinId = 'sui'; // CoinGecko id

const [markets, news, narrative] = await Promise.all([
  agent.pay({
    url: 'https://mpp.t2000.ai/coingecko/v1/markets',
    method: 'POST',
    body: JSON.stringify({ vs_currency: 'usd', ids: coinId, sparkline: true }),
  }),
  agent.pay({
    url: 'https://mpp.t2000.ai/newsapi/v1/search',
    method: 'POST',
    body: JSON.stringify({ q: 'Sui blockchain', pageSize: 5, sortBy: 'publishedAt' }),
  }),
  agent.pay({
    url: 'https://mpp.t2000.ai/perplexity/v1/chat/completions',
    method: 'POST',
    body: JSON.stringify({
      model: 'sonar',
      messages: [{ role: 'user', content: 'What is the current narrative and recent developments for the Sui blockchain? Cite sources.' }],
    }),
  }),
]);

const take = await agent.pay({
  url: 'https://mpp.t2000.ai/anthropic/v1/messages',
  method: 'POST',
  headers: { 'anthropic-version': '2023-06-01' },
  body: JSON.stringify({
    model: 'claude-sonnet-4-5',
    max_tokens: 600,
    messages: [{
      role: 'user',
      content:
        `Write a 200-word deep-dive on ${coinId}. Lead with price + 24h/7d move, then the narrative ` +
        `and key catalysts. Keep the citations from the research.\n\n` +
        `MARKETS: ${JSON.stringify(markets.body)}\n\nNEWS: ${JSON.stringify(news.body)}\n\nNARRATIVE: ${JSON.stringify(narrative.body)}`,
    }],
  }),
});

console.log((take.body as { content: { text: string }[] }).content[0].text);
```

***

## Expected output

```
4 calls · ~$0.08 · ~8s · 0 taps
SUI deep-dive: market data + narrative + cited take
```

***

## Extend it

* Add **AlphaVantage** (`/alphavantage/v1/quote`) to fold in a correlated equity (e.g. COIN, MSTR) for a cross-asset view
* Swap **Perplexity** for **Exa** (`/exa/v1/search`) + **Firecrawl** (`/firecrawl/v1/scrape`) to read primary sources directly
* Run it across your watchlist in parallel and have Claude rank by momentum
* Schedule it as a morning cron and **Pushover** (`/pushover/v1/push`) the take to your phone