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

# Quickstart

> Create an API key and run one hosted fetch.

## 1. Sign up

Create a free account at [hermai.ai/register](https://hermai.ai/register). The registry is free to browse.

## 2. Create an API key

Go to **Dashboard > API Keys** and create a new key. Copy it once you create it. You need it for authenticated requests.

```
hm_sk_abc123...
```

Some registry browsing requests work anonymously, but hosted fetch requires an API key.

## 3. Run hosted fetch

Call a registered endpoint through Hermai:

```bash theme={null}
export HERMAI_API_KEY="hm_sk_abc123..."

curl -sS -X POST "https://api.hermai.ai/v1/fetch" \
  -H "Authorization: Bearer ${HERMAI_API_KEY}" \
  -H "Content-Type: application/json" \
  -d '{
    "site": "semanticscholar.org",
    "endpoint": "get_paper",
    "params": {
      "paper_id": "ARXIV:1706.03762",
      "fields": "paperId,externalIds,title,authors,venue,year,referenceCount,citationCount,openAccessPdf"
    }
  }'
```

Response:

```json theme={null}
{
  "success": true,
  "data": {
    "paperId": "204e3073870fae3d05bcbc2f6a8e263d9b72e776",
    "title": "Attention is All you Need",
    "year": 2017
  },
  "meta": {
    "site": "semanticscholar.org",
    "endpoint": "get_paper",
    "method": "GET",
    "source": "gateway",
    "latency_ms": 1234,
    "credits_used": 1,
    "credits_remaining": 999,
    "cached": false
  }
}
```

## 4. Browse available schemas

Explore what's available at [hermai.ai/schemas](https://hermai.ai/schemas), or via the API:

```bash theme={null}
# List all schemas
curl "https://api.hermai.ai/v1/schemas"

# Get one public schema card
curl "https://api.hermai.ai/v1/schemas/semanticscholar.org"
```

Public schema cards show the site, description, endpoint names, methods, purposes, and user-facing readiness signals.

## 5. Pull a package or self execute

If you want your own agent or server to execute requests directly, pull the catalog or full package with an intent:

```bash theme={null}
curl -H "Authorization: Bearer ${HERMAI_API_KEY}" \
     -H "X-Hermai-Intent: looking up product endpoints on allbirds.com to build a price comparison agent" \
     "https://api.hermai.ai/v1/catalog/allbirds.com"

curl -H "Authorization: Bearer ${HERMAI_API_KEY}" \
     -H "X-Hermai-Intent: downloading Semantic Scholar metadata routes for citation enrichment" \
     "https://api.hermai.ai/v1/schemas/semanticscholar.org/package"
```

The catalog and package routes return the endpoint recipe. `POST /v1/fetch` executes a registered endpoint through Hermai.

## 6. Contribute a schema

Found a site that's missing? Document its endpoints with the CLI and push the schema:

```bash theme={null}
# Install the CLI
go install github.com/hermai-ai/hermai-cli/cmd/hermai@latest

# Classify the site + find its public paths
hermai detect https://example.com
hermai wellknown example.com

# Pull embedded data from a detail page
hermai probe --body https://example.com/products/123 | hermai extract

# Capture XHR traffic for dynamic pages (search, cart, etc.)
hermai intercept https://example.com/search?q=test

# Author schema.json (see /concepts/schemas), then push
hermai registry push schema.json
```

<Tip>
  **Using Claude Code, Codex, or Cursor?** Run `npx skills add hermai-ai/hermai-skills` to add the Hermai registry workflow. See the [hermai-skills repo](https://github.com/hermai-ai/hermai-skills) for details.
</Tip>

You earn **+50 credits per site** for contributions. Credits are tracked in your dashboard and can be used for hosted execution where available.

## Next steps

* [How Hermai works](/concepts/how-hermai-works): the runtime model
* [Hosted fetch](/concepts/hosted-fetch): run registered endpoints through Hermai
* [Schemas](/concepts/schemas): format, verification, contributing
* [Catalog](/concepts/catalog): the lookup API in detail
* [API keys](/concepts/api-keys): scopes and rate limits