Connect VolunteerBadge to Claude

You can run FCRA-compliant volunteer background checks directly from Claude — Claude Desktop, Claude.ai, or Claude Code — using VolunteerBadge's hosted connector. Nothing to install on a server; you connect one URL.

The connection URL

https://www.volunteerbadge.com/api/mcp

Three hard rules (built in)

Never type a Social Security Number into chat. The subject enters their own SSN on a secure VolunteerBadge page; it's optional and only enriches address history.
Get authorization first. By default Claude sends the person a secure link to authorize before any check runs.
Records found ≠ reject. The law requires a pre-adverse-action notice with a copy of the report before any decision. Notices must be sent from the VolunteerBadge dashboard (Compliance center or the volunteer/direct-check workflow) — not from Claude chat.

Option 1 — Claude Desktop (one-click)

Open Claude Desktop → Settings → Connectors.
Click “Add custom connector”. Name it VolunteerBadge and paste the URL above.
Click Connect → log into VolunteerBadge in the popup → pick your organization → Authorize.
Done — the VolunteerBadge tools appear. Ask: “What's my VolunteerBadge credit balance?”

Option 2 — Claude.ai (web)

Same as Claude Desktop: Settings → Connectors → Add custom connector, paste the URL, click Connect, and complete the VolunteerBadge sign-in + authorize. (Custom connectors require a Pro, Max, Team, or Enterprise plan.)

Option 3 — Claude Code (one command)

First create an API key: VolunteerBadge dashboard → Settings → API → agree to the End-User Agreement → Create key. Then:

claude mcp add --transport http volunteerbadge \
  https://www.volunteerbadge.com/api/mcp \
  --header "Authorization: Bearer vb_live_xxxxxxxx"

Any MCP client works the same way — use the URL with an Authorization: Bearer <key> header.

Scenarios you can run

Once connected, just ask in plain English — Claude picks the right VolunteerBadge tool, keeps you consent-first, and never needs a Social Security Number in chat:

Ask Claude…	What happens
“What’s my VolunteerBadge credit balance?”	Reads your remaining checks so you know what’s available before ordering.
“Send a background-check authorization to Maria Lopez at maria@example.org, then run her check once she signs.”	Emails the subject a secure consent link; the $5 check runs automatically after she authorizes — no SSN ever touches the chat.
“Order checks for everyone on my approved roster who hasn’t been screened in the last year.”	Lists your volunteers, finds the ones due, and queues the whole batch from a single prompt.
“Which of my pending checks have come back, and what do they say?”	Pulls the current status of each check and summarizes the results (clear vs. needs review).
“A check came back with a record — what does it mean and what do I have to do?”	Explains the result and the FCRA pre-adverse → wait → final steps. Directs you to Compliance or the adverse-action drawer to send notices by email — it won’t generate or send notices from chat.
“Do volunteers in Florida need to be re-screened every year?”	Answers FCRA and screening-policy questions in plain English.

Building an automated workflow instead of chatting? The same actions are available over the REST API & webhooks.

The Skill (optional)

The connector's tools already enforce the rules above. For extra workflow guidance, install the VolunteerBadge Skill: download the Skill (.zip) and upload it in your Claude app's Skills / Capabilities area (availability varies by plan and version).

First check blocked? That's expected.

Two one-time setup steps can block a first check — both quick to clear:

“Your organization must re-accept the CRA End-User Agreement” — an owner/admin accepts the current version in Settings → API, then retry.
ID_VERIFICATION_REQUIRED on an instant check (request_instant_check) — your organization's owner completes a free, one-time identity verification in Settings → Account (“Verify my identity”), then retry. Consent-first checks (request_background_check) aren't affected.