Scraper resource

The scraper executor is a native capability compiled into the kdeps binary. It fetches a URL and returns the text content, with optional CSS selector filtering.

Native scraper: is HTML text. For PDFs, .docx, .xlsx, and images, install the registry component of the same name (kdeps registry install scraper) and call it with component:. They are not the same thing.

Where it runs

Both workflow mode and agent mode. In workflow mode it executes as a DAG step. In agent mode, the workflow containing this resource runs as a single callable tool.

Configuration

# resources/fetch.yaml
scraper:
  url: "https://example.com"     # required
  selector: "article.content"    # optional CSS selector
  timeout: 30                    # seconds (default: 30)

Field	Type	Required	Default	Description
url	string	yes	-	URL to fetch
selector	string	no	-	CSS selector to scope extraction
timeout	integer	no	30	Request timeout in seconds

Output

Key	Type	Description
content	string	Extracted text (full body or selector match)
url	string	The URL that was fetched
status	integer	HTTP status code
json	string	Full result as a JSON string

Access fields with output('actionId').content etc.

Examples

Fetch a page and summarize

# resources/fetch.yaml
actionId: fetch
scraper:
  url: "{{ get('url') }}"

---
actionId: summarize
requires: [fetch]
chat:
  model: llama3.2:1b
  prompt: "Summarize: {{ output('fetch').content }}"
apiResponse:
  response: "{{ output('summarize') }}"

Extract with CSS selector

# resources/fetch-article.yaml
actionId: fetchArticle
scraper:
  url: "https://news.example.com/article"
  selector: "article.body"

Error handling

Use onError to handle unreachable URLs gracefully:

# resources/example.yaml
scraper:
  url: "https://example.com"
onError:
  action: continue
  fallback: ""

---

Need more? For PDF extraction, OCR, and document types (.docx, .xlsx), install the component:

kdeps registry install scraper

See also

• Embedding resource - store and search scraped content
• LLM resource - feed scraped content into prompts
• HTTP client - lower-level HTTP with auth, headers, caching