WEB DATA & EXTRACTION · FOR AI AGENTS
Wappalyzer
Retrieve website technology and company signals through a maintained index, API or hosted MCP.
Wappalyzer supplies technology detections and optional company enrichment for website research. Agents can choose indexed results or a live lookup, then use the structured findings to qualify an account or assess a possible integration.
Your integration checklist
- Capabilities
- What you need from your owner
- Use a Business account with an existing API key. Connect the hosted MCP endpoint using the account sign-in flow, or configure a server-side REST client.
- Authentication
- REST uses x-api-key. Hosted MCP at https://mcp.wappalyzer.com/mcp uses OAuth sign-in and reads the account’s existing API key; the client does not need a pasted key.
- Your first task
- Look up a known company domain and distinguish cached detections from a current live observation before choosing an integration.
- Billing & payment
- API calls spend account credits. Lookup cost depends on cached versus live recursive execution; the free website allowance is not Business API access. Review credit balance and expiration before a batch.
- Agent payment support
- Not verified. Programmatic usage does not establish autonomous checkout support.
- Operational limits
- The hosted MCP server is read-only and forces recursive=false for synchronous website lookup. REST recursive live scans complete asynchronously. Detected technology does not establish an active contract or exact deployment configuration.
ORIGINAL AI AGENT STORE INTEGRATION PLAYBOOK
Choose the lookup mode before interpreting an empty technology result
Editorial recommendation · Reviewed 2026-11-14
When to choose this service
Use Wappalyzer for normalized website technology and enrichment data. A direct page inspection may suffice when only one visible feature matters.
The maintained index, detection rules and hosted crawling supply external observations. Ordinary code can inspect one page but does not recreate that accumulated coverage.
Suggested workflow
- Normalize the intended URL and select cached or live lookup based on freshness needs. Check the account balance.
- For hosted MCP, retrieve a synchronous single-site result. For recursive REST, persist the pending request and wait for its callback or documented follow-up.
- Keep observation context with the result and verify consequential technology claims against the current site before recommending changes.
Acceptance checks before real work
- Pending recursive work is not evidence that the site has no technologies.
- Hosted MCP and recursive REST have different completion behavior.
- A free website lookup quota does not establish API entitlement.
If the workflow fails
If a result is incomplete, check mode and job state before paying for another crawl. For callbacks, verify the documented signature and reconcile the original request.