Spis/Docs

CLI command

spis crawl-web

Crawl real browser products by submitting a Weles plan on a pinned Stado host.

Invocation#

bash
spis crawl-web <catalog> --host <target> --admission-url <url> [--record <slug>] [--accounts <file>] [--wait-seconds <n>]

Crawl real browser products by submitting a Weles plan on a pinned Stado host.

Inputs and options#

  • <catalog> is required and must be one of the eight browser catalogs named by the refusal below.
  • --host <target> and --admission-url <url> are required in coordinator mode; the admission URL must be a valid loopback URL.
  • --record <slug> selects one record; --accounts <file> supplies account bindings.
  • --wait-seconds <n> defaults to 7200 and must be 30–86400.
  • --worker, --plan, and --plan-base64 are worker-facing options.

Output and state effects#

The coordinator writes a wisent.web-crawl-plan.v1 to target/spis-crawl-plans/ and submits it with the exact source revision. The worker enqueues Weles tasks, waits for job results, prints a wisent.web-crawl-run.v1 JSON report, and errors if any Weles crawl job failed.

Refusals#

These messages are emitted by the current Rust implementation:

  • catalog is required
  • crawl-web accepts web-app-examples, dashboard-console-examples, onboarding-auth-examples, app-store-listing-examples, design-system-examples, report-evidence-examples, pricing-page-examples, landing-page-examples
  • --admission-url is required and --host is required; web crawls execute through Weles on a pinned Stado host
  • --wait-seconds must be 30..86400
  • Unknown flags and extra positionals are refused; coordinator mode refuses worker plans with --plan and --plan-base64 are worker-only.