Search test library by skills or roles
⌘ K
Model Context Protocol

Bring Adaface into your AI assistant with MCP

Connect Adaface to Claude, Cursor and any MCP-compatible AI client. Ask your assistant to list assessments, check candidate scores, and send assessment invites, all in plain language and without leaving your chat.

See how it works
Adaface MCP integration

What is the Adaface MCP server?

The Model Context Protocol (MCP) is an open standard that lets AI assistants securely connect to external tools. The Adaface MCP server exposes your assessment workflows as tools your AI client can call, authenticated with your Adaface API key. No glue code and no scraping, just connect and go.

What your assistant can do


Read

Available by default, safe read-only access.

list_tests: browse your assessments
get_test_details: duration, cut-off, question mix
get_test_status: a candidate's score & scorecard
list_candidates: a candidate's assessments
list_interview_ids: page through past interviews
get_credits_usage: your credit balance
list_test_requests: your test requests and status
search_test_library: Adaface's ready-made tests
get_library_test_details: details of a library test

Read & write

Opt-in mode that can act on your account.

send_test: email an assessment to a candidate
resend_invite: re-send an invite
send_reminder: nudge a candidate
cancel_test: cancel a pending invite
shortlist_or_archive: triage candidates
create_test_share_link: make a self-serve test link
create_ready_to_go_test: spin up a test
request_custom_test: request a custom-built test

You choose the access level


Read only

The default. Your assistant can look up tests, candidates and scores, but cannot change anything or contact candidates. Perfect for reporting and status checks.

Read & write

Opt in when you want your assistant to take action, including sending assessment invites to candidates. You can switch modes any time from Settings → Integrations → MCP.

Install in under a minute

Enable MCP from Settings → Integrations → MCP in your Adaface dashboard, then add this to your MCP client's configuration using your Adaface API key.

{
  "mcpServers": {
    "adaface": {
      "url": "https://api.adaface.com/app/api/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_ADAFACE_API_KEY"
      }
    }
  }
}

Read the documentation →

Secure by design

Every request is authenticated with your Adaface API key and scoped to your account. Write actions are off unless you explicitly enable read & write mode, and all MCP calls are rate-limited to keep your account safe.

customers across world
Join 1200+ companies in 80+ countries.
Try the most candidate friendly skills assessment tool today.
g2 badges