This feature is currently in beta and requires passing the parallel-beta: source-policy-2025-07-03 header.

Source Policy is available for both the Parallel Task API and Search API - giving you granular control over the sources Parallel processors has access to. This feature enables you to allow or disallow specific domains to tailor search relevance.

Configuration

Configure domain filtering using these parameters:

ParameterTypeDescription
include_domainsArray of stringsList of domains to allow. Only sources from these domains will be included in results.
exclude_domainsArray of stringsList of domains to block. Sources from these domains will be excluded from results.

When you specify a domain like example.com, it will match the base domain and all its subdomains like example.com, www.example.com, blog.example.com, api.example.com

You can specify up to 10 domains per request. Exceeding this limit will result in an error.
curl -X POST "https://api.parallel.ai/v1/tasks/runs" \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -H "parallel-beta: source-policy-2025-07-03" \
  -d '{
      "task_spec": {
        "output_schema": "How many employees are there in the company"
      },
      "input": "Parallel Web Systems",
      "processor": "core",
      "source_policy": {
        "include_domains": ["linkedin.com"]
      }
    }'

Best Practices

  • Use either include_domains or exclude_domains. Do not use both in one query. If you set both parameters, only domains in include_domains will be considered.
  • List each domain in its apex form (example.com) and omit any schemes (http://, https://) or subdomain prefixes such as (www.).
  • Even though there is a maximum limit of 10 domains, carefully using specific and targeted domains will give better results.