Website operations

Check Website Data

Check if a company has website data (crawled and enriched). Returns True if website exists and has been crawled.

🔒 Requires Authorization: Bearer ll_… (or X-API-Key).

Path parameters

Example request

curl -X GET "https://api.lunalift.ai/website/has-data/COMPANY_ID" \
  -H "Authorization: Bearer ll_your_api_key"

Response · 200 OK

Returns a JSON object on success.

Trigger Website Enrichment

Trigger website enrichment workflow for a company's website. This endpoint initiates the website crawling and enrichment process. The workflow runs asynchronously in the background. If domain is not provided, it will be fetched from the company record.

🔒 Requires Authorization: Bearer ll_… (or X-API-Key).

Request body· TriggerEnrichmentRequest

Example request

curl -X POST "https://api.lunalift.ai/website/trigger-enrichment" \
  -H "Authorization: Bearer ll_your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
  "company_id": "string"
}'

Response · 200 OK

Recrawl Page

Re-crawl a single page — FAST crawl only, no enrichment side effects. Routes the one page onto the same fast RabbitMQ crawl the nightly ``customer_page_crawl_wf`` uses (``WebFetchService.queue_customer_page_for_fetch`` → ``url_fetch_queue`` → rust parser → ``result_queue`` → ``AsyncBatchedConsumer`` → ``promote_slim_to_webpage``, which advances ``website.last_crawl_at``). The sync icon deliberately does NOT run the schema / KB / llms.txt enrichment DAG: schema regen has its own ``/regen-page`` entrypoint and llms.txt its own ``/generate-llms-txt`` entrypoint. ``company_id`` is a PATH param (matching issues.py's ``/companies/{company_id}/...`` convention). Endpoint URL + ``RecrawlPageResponse`` shape are unchanged (frontend untouched).

🔒 Requires Authorization: Bearer ll_… (or X-API-Key).

Path parameters

Request body· RecrawlPageRequest

Example request

curl -X POST "https://api.lunalift.ai/website/COMPANY_ID/recrawl-page" \
  -H "Authorization: Bearer ll_your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
  "url": "string"
}'

Response · 200 OK

Regen Page Schema

Enqueue a per-page schema regen for ONE page (``url_path`` is a typed query param — never a URL-in-body / ``Dict[str, Any]``). Publishes a single job to ``schema_regen_queue`` (same RabbitMQ publish path the crawl fan-out uses) and returns immediately — regen never runs inline on the request. The consumer calls ``schema_regen_service.regenerate_page_schema`` which rebuilds ONLY this page's served ``.schema.json`` from the base ⊕ overlay KB (WS1 single read path) and advances its ``dateModified``.

🔒 Requires Authorization: Bearer ll_… (or X-API-Key).

Path parameters

Query parameters

Example request

curl -X POST "https://api.lunalift.ai/website/COMPANY_ID/regen-page?url_path=VALUE" \
  -H "Authorization: Bearer ll_your_api_key"

Response · 200 OK

Regen Entity Pages

Bounded overlay-edit fan-out: an edited ``canonical_id`` → reverse-lookup the pages that carry it (WS3 ``parent_canonical_id`` index) → enqueue a per-page regen for ONLY those pages. Never re-enriches the whole storage_key for one edit. ``canonical_id`` is a typed query param. Returns immediately with the pages enqueued.

🔒 Requires Authorization: Bearer ll_… (or X-API-Key).

Path parameters

Query parameters

Example request

curl -X POST "https://api.lunalift.ai/website/COMPANY_ID/regen-entity?canonical_id=VALUE" \
  -H "Authorization: Bearer ll_your_api_key"

Response · 200 OK

Regen All Schema

Enqueue a schema regen for ALL of the company's NON-SYSTEM pages. Mirrors ``/regen-page`` (same ``_authorize_company`` guard, PATH-scoped ``company_id``, same ``schema_regen_queue`` publish path) but fans out one job per non-system page instead of one. Bounded fan-out — the endpoint returns immediately with the count; regen never runs inline. Targets EVERY non-SYSTEM page including schema-LESS ones (the point of bulk regen is to give schema to pages that lack it); only system pages (legal/checkout/account/search/error) are skipped, via the SHARED classifier in ``schema_regen_service``.

🔒 Requires Authorization: Bearer ll_… (or X-API-Key).

Path parameters

Example request

curl -X POST "https://api.lunalift.ai/website/COMPANY_ID/regen-all" \
  -H "Authorization: Bearer ll_your_api_key"

Response · 200 OK

Generate Llms Txt Endpoint

Regenerate + serve this company's ``llms.txt`` — its OWN entrypoint. llms.txt generation was lifted out of the ``website_enrichment_wf`` DAG (it used to be the terminal ``generate_llms_txt`` task chained off ``update_schema_descriptions``). It now runs on-demand here, decoupled from crawl and schema regen. Reuses the identical service path the DAG task used (``WebsiteService.generate_and_upload_llms_txt`` → ``generate_llms_txt``), so WS2's tombstone suppression (``read_tombstoned_canonical_ids``) still applies. No crawl, no schema regen — llms.txt only.

🔒 Requires Authorization: Bearer ll_… (or X-API-Key).

Path parameters

Example request

curl -X POST "https://api.lunalift.ai/website/COMPANY_ID/generate-llms-txt" \
  -H "Authorization: Bearer ll_your_api_key"

Response · 200 OK