Website Technology Checker — Wappalyzer & BuiltWith Alternative

Find out what a website is built with. Check up to 500 public website URLs per run. Discover CMS, ecommerce platforms, analytics, marketing tools, JavaScript frameworks, hosting, CDN and visible security technologies. Export JSON, CSV or Excel for prospect research, CRM enrichment and website comparisons.

Start with three choices

1. Websites to check: paste domains or page URLs, one per line.
2. What would you like to check? Discover the technology stack, check specific tools, or compare technologies across pages.
3. How much should we check? Choose Fast for a basic check or Extended for broader coverage. Extended can take longer and increase platform usage.

For specific tools, enter names such as Shopify, WordPress or Google Analytics. Optional filters select which websites enter the results. Advanced settings control additional checks, page limits and optional traffic estimates.

{
  "urls": ["wordpress.org"],
  "scan_goal": "check_tools",
  "scan_profile": "fast",
  "target_technologies": ["Shopify", "WordPress"]
}

Use scan_goal: "overview" for an overview or "compare_pages" to compare up to three pages per website. Unknown tool names give an input error. Technology names ignore letter case.

Read your results

Open Check report for findings, unavailable inputs and recommended next steps. Check answers provides a compact dataset view. Check report — JSON for integrations provides the report for automation.

Status	Meaning
signal_found	A technology signal was observed. This does not verify that the tool ran or sent an event.
not_observed	No matching signal was observed within the checked scope. This does not prove the tool is absent.
inconclusive	A required check could not be completed.

Each delivered website has one dataset row. Use tech_names for a technology list, technologies for categories and available versions, or flat category columns for spreadsheets. input_url and checked_at help join results to your source list. page_checks, coverage_status and next_step explain incomplete checks. detection_sources identifies a page or additional check associated with a finding; it is not a confidence score.

rendered_check_status and rendered_added_count describe the optional additional check and its new findings. Such checks may be unavailable or rate-limited; existing findings are retained. Fast skips additional checks. Extended enables them unless disabled with include_rendered_check: false. max_rendered_sites limits these checks per run.

Dataset views focus on Ecommerce & CMS, Analytics & Marketing, Frontend & Frameworks, Infrastructure, Security or Detection Details. JSON preserves nested data; CSV and Excel provide spreadsheet exports.

Optional Similarweb traffic estimates add estimated visits, rank, engagement and acquisition channels. Enable include_similarweb to request them. Estimates may be unavailable and are not first-party analytics.

The first-run example uses WordPress.org. Some websites refuse cloud requests: Allbirds returned HTTP 403 in the October 2026 checks. These inputs produce diagnostics rather than a guessed technology stack. Optional rendering supplements a readable sparse page; it is not a bypass for a blocked base request.

Filters, diagnostics and retries

Set technologies_filter to names such as Shopify or categories_filter to categories such as Ecommerce. A website must match at least one name in each supplied filter. Matching rows retain their full detected stack. Tool checks alone do not filter websites.

Equivalent homepage URLs are checked once. Different paths and query strings remain separate inputs. Invalid, duplicate, filtered, blocked, unreachable and zero-signal inputs appear in SCAN_ERRORS and the report.

SUMMARY contains run counts. RETRY_INPUT preserves your options and lists inputs affected by temporary failures. Check that its urls list is non-empty before another run.

SUMMARY.status is complete, partial, empty or source_unavailable. If every unique input is invalid or unavailable, the run fails after saving diagnostics. A readable zero-signal page or a filter excluding all sites is a valid empty result. Partial runs retain useful rows. A budget stop preserves remaining inputs in RETRY_INPUT; no source request is made when the event budget cannot cover a result.

Pricing

$0.0015 per delivered website — $1.50 per 1,000 results, plus startup and Apify platform/proxy usage paid by the user. The startup event costs $0.00005 per GB, with a minimum of one event; at the default 4 GB allocation it is $0.0002. Check the Pricing tab for current charges.

Skipped and failed inputs incur no result event, but their resource usage is billed under your Apify plan. Additional checks can increase usage costs. Apify's maximum total charge setting limits event spending.

Coverage and automation

Results describe observable website signals at the time of checking. Websites can change; versions are included only when observable. A missing tool is not proof of absence. Page comparison covers up to three pages, not the entire website. body_truncated indicates a partly checked page. This service does not verify event delivery, consent behavior, every browser interaction or email-provider configuration.

Run the Actor through Apify's API with the same JSON input, then read its dataset and report. Older saved input options remain supported; new integrations should use the current input schema and select a check profile explicitly.

REST API, MCP and agent contract

Use this Actor when you already have public domains or page URLs and need a current technology scan. Use Tech Stack Leads & Lookup for searching an indexed database. This Actor does not discover prospect URLs, validate tracking events, test cookie consent, inspect all browser network activity or monitor changes across runs.

Start asynchronously with POST https://api.apify.com/v2/acts/apivault_labs~website-tech-detector/runs. Authenticate with Authorization: Bearer <APIFY_TOKEN> and send the JSON example above. Poll GET /v2/actor-runs/{runId} to a terminal status; read GET /v2/datasets/{defaultDatasetId}/items and GET /v2/key-value-stores/{defaultKeyValueStoreId}/records/SUMMARY. Read SCAN_ERRORS, CHECK_REPORT and RETRY_INPUT even if the run failed. Add ?format=csv&clean=true to the dataset items endpoint for CSV. Keep credentials in your client configuration.

Connect an authenticated MCP client to:

https://mcp.apify.com/?tools=apivault_labs/website-tech-detector

Discover tools with tools/list, then invoke apivault_labs--website-tech-detector with the same input fields. A short waitSecs can return before completion: follow the returned run and storage IDs. Use the discovered get-dataset-items and get-key-value-store-record tools to read results and diagnostics. Do not rerun solely because the waiting interval elapsed.

Agents should explicitly choose scan_profile: "fast" for a quick supplied-page scan. Use "extended" for up to three eligible same-host pages and optional rendered enrichment; set include_rendered_check: false if you need HTTP-only comparison. Request include_similarweb: true only when third-party traffic estimates are wanted. Legacy requests omitting a profile retain legacy enrichment defaults, so specify a profile for predictable time and cost.

For each row verify tech_count == technologies.length, checked_at, final_url and coverage_status. Technology fingerprints can include rule implications; detection_sources identifies the checked page or additional check, not exact matching evidence, a direct/inferred classification or independent proof of backend software. Report not_observed as inconclusive about absence. Rendered checks can return blocked, unavailable, rate_limited, run_limit or time_budget_reached; these preserve base findings and mark coverage partial. Never label these as full browser coverage.

For reliable automation, store the original input alongside the run ID and retry only non-empty RETRY_INPUT.urls after the stated temporary failure or budget restriction is resolved. One delivered row is one tech-detection event; skipped inputs and reports incur no result event. Maximum total charge is an event budget; separate platform/proxy usage can still apply.

The service has its own input and output format and is not affiliated with Wappalyzer or BuiltWith. For finding websites in an indexed technology database, see Tech Stack Leads & Lookup.