# Map the paid agent API market in one prompt

Before building on paid APIs, it helps to know what's out there. One call lists every category Hagglyai compares, how many paid endpoints each has and the cheapest price.

## 1. Add Hagglyai to Claude

### In the Claude app (chat box)

1. In Claude, go to **Customize > Connectors**.
2. Click **+ Add**, then **Add custom connector**.
3. Name it **Hagglyai** and enter the server URL `https://hagglyai.com/mcp`, then click **Continue** twice.
4. Under Authentication choose **No sign in** (Hagglyai needs no account or key), then click **Add**.
5. In a chat, click **+** in the lower left, choose **Connectors** and turn **Hagglyai** on.
6. Type your question in the chat box. If Claude asks before using a Hagglyai tool, allow it.

Pro and Max add connectors themselves; Free plans can have one custom connector; on Team and Enterprise an owner adds it and members click Connect. Once added it also works in the desktop and mobile apps.

### In Claude Code (one line)

```sh
claude mcp add --transport http haggly https://hagglyai.com/mcp
```

Any other MCP client: add the server `https://hagglyai.com/mcp` (Streamable HTTP, no key). More at https://hagglyai.com/for-agents.

## 2. Ask

> What kinds of paid APIs can my agent use, and how cheap do they get?

Claude calls `list_categories` with:

```json
{}
```

## 3. The answer (live, as of 2026-10-27 11:13 UTC)

# Hagglyai categories

| Category | id | Paid endpoints | MCP servers | Cheapest/unit |
|---|---|---|---|---|
| Web search | `web-search` | 532 | 117 | $0.0003 |
| Web page extraction | `web-extraction` | 1484 | 272 | $0.0001 |
| LLM inference | `llm-inference` | 931 | 43 | $0.0000078 |
| Crypto prices & market data | `crypto-prices` | 1489 | 50 | $0.000002 |
| On-chain data | `onchain-data` | 1388 | 6 | $0.0001 |
| People & company enrichment | `enrichment` | 1000 | 244 | $0.0003 |
| Weather | `weather` | 6390 | 140 | $0.0002 |
| News | `news` | 556 | 93 | $0.0005 |
| Social media data | `social-media` | 948 | 290 | $0.00002 |
| Job listings | `jobs` | 424 | 88 | $0.000033 |
| Products & prices | `ecommerce` | 254 | 142 | $0.00005 |
| Maps & places | `maps-places` | 132 | 38 | $0.0002 |
| Translation | `translation` | 510 | 83 | $0.001 |
| Email, SMS & messaging | `messaging` | 49 | 118 | $0.001 |
| Flights & travel | `travel` | 173 | 132 | $0.0002 |
| Stocks & forex | `stocks` | 1257 | 257 | $0.000002 |
| PDF generation | `pdf-generation` | 80 | 62 | $0.002 |
| Captcha solving | `captcha-solving` | 10 | 5 | $0.001 |

Compare one: https://hagglyai.com/compare/{id}.md

## Sub-categories of `web-search` (sellers in brackets)

- Search engine: `google` (76), `exa` (26), `brave` (16), `bing` (12), `tavily` (12), `duckduckgo` (9), `perplexity` (6)
- Task: `results` (144), `answers` (70), `news` (67), `semantic` (25)

## Sub-categories of `web-extraction` (sellers in brackets)

- Task: `markdown` (318), `metadata` (168), `crawl` (127), `browser` (117), `robots` (94), `structured` (82), `documents` (79), `screenshots` (16)

## Sub-categories of `llm-inference` (sellers in brackets)

- Model: `openai` (52), `meta-llama` (32), `qwen` (30), `google` (26), `anthropic` (23), `deepseek` (22), `mistral` (8), `xai` (8)
- Task: `chat` (126), `reasoning` (21), `embeddings` (10), `speech` (9), `images` (4)

## Sub-categories of `crypto-prices` (sellers in brackets)

- Coin: `ethereum` (138), `bitcoin` (134), `solana` (109), `stablecoins` (41)
- Task: `spot` (162), `market` (128), `dex` (96), `ohlcv` (94), `derivatives` (27)

## Sub-categories of `onchain-data` (sellers in brackets)

- Chain: `base` (182), `ethereum` (90), `polygon` (62), `arbitrum` (59), `optimism` (56), `solana` (39), `bnb` (21), `bitcoin` (17)
- Task: `blocks` (164), `balances` (158), `transactions` (149), `tokens` (133), `rpc` (79), `risk` (31)

## Sub-categories of `enrichment` (sellers in brackets)

- Task: `company` (166), `registry` (76), `people` (73), `email` (54), `phone` (44), `domain` (22)

## Sub-categories of `weather` (sellers in brackets)

- Data source: `noaa` (48), `open-meteo` (40), `met-norway` (27)
- Task: `forecast` (129), `current` (123), `alerts` (49), `historical` (19), `air-quality` (18), `marine` (11)

## Sub-categories of `news` (sellers in brackets)

- Task: `headlines` (167), `crypto` (103), `topic-search` (78), `sentiment` (48), `tech` (33)

## Sub-categories of `social-media` (sellers in brackets)

- Platform: `youtube` (53), `x-twitter` (52), `instagram` (25), `tiktok` (23), `reddit` (21), `telegram` (18), `facebook` (15), `bluesky` (12), `linkedin` (10), `threads` (9), `pinterest` (4)
- Task: `posts` (104), `search` (60), `profiles` (48), `transcripts` (38), `comments` (32), `analytics` (22), `publish` (5)

## Sub-categories of `jobs` (sellers in brackets)

- Source: `ats` (14), `linkedin` (8), `indeed` (5)
- Task: `search` (52), `companies` (22), `salary` (17), `remote` (15)

## Sub-categories of `ecommerce` (sellers in brackets)

- Marketplace: `amazon` (22), `shopify` (14), `ebay` (9), `google-shopping` (9), `walmart` (5), `etsy` (3), `aliexpress` (3)
- Task: `products` (39), `reviews` (23), `sellers` (23), `prices` (22)

## Sub-categories of `maps-places` (sellers in brackets)

- Source: `google-maps` (26), `tripadvisor` (4), `openstreetmap` (3)
- Task: `places` (32), `geocoding` (24), `reviews` (19), `routing` (7)

## Sub-categories of `translation` (sellers in brackets)

- Engine: `llm` (8), `libretranslate` (3)
- Task: `text` (74), `detect` (36), `websites` (25), `documents` (14), `speech` (12), `subtitles` (8)

## Sub-categories of `messaging` (sellers in brackets)

- Channel: `email` (18), `sms` (9), `voice` (6)
- Task: `send` (24), `receive` (11)

## Sub-categories of `travel` (sellers in brackets)

- Source: `google` (7), `airbnb` (5), `booking` (3)
- Task: `airports` (28), `flights` (22), `status` (21), `hotels` (9), `reviews` (8), `ground` (3)

## Sub-categories of `stocks` (sellers in brackets)

- Data source: `sec-edgar` (128), `yahoo-finance` (13), `tradingview` (5), `alpha-vantage` (4), `finnhub` (3)
- Task: `forex` (184), `filings` (134), `quotes` (125), `history` (73), `fundamentals` (71), `earnings` (51), `screeners` (16)

## Sub-categories of `pdf-generation` (sellers in brackets)

- Input: `url` (21), `templates` (13), `html` (12), `markdown` (8), `images` (6), `office` (3)
- Task: `invoices` (15), `reports` (12), `contracts` (5), `screenshots` (4)

## Sub-categories of `captcha-solving` (sellers in brackets)

- Captcha type: `turnstile` (3)

Source: hagglyai.com · as of 2026-10-27 11:13 UTC

## 4. How to read it

- Each category lists its paid endpoints and the cheapest price per unit.
- The sub-category ids (platforms and tasks) narrow a later comparison, e.g. Instagram comments or YouTube transcripts.
- Use a category id with `compare_category` or `most_popular` for the detail.

More guides: https://hagglyai.com/guides.md