Skip to main content

Claude Key Setup

Anthropic's Claude models can power the Activity Summarizer, Extractor, and ToolboxAI tools. To use Claude, you'll need an API key from Anthropic and a few dollars of credit on your Anthropic account.

1. Create an Anthropic account​

If you don't already have one, sign up at console.anthropic.com.

2. Add billing to your Anthropic account​

  1. In the Anthropic Console, go to Settings → Billing.
  2. Add a payment method and add credit to your account. ToolboxAI calls the Anthropic API on a pay-as-you-go basis, so your account needs an active payment method before any requests will succeed.

-- IMAGE COMING SOON --

3. Create an API key​

  1. In the Anthropic Console, go to Settings → API Keys.
  2. Click Create Key.
  3. Give it a name (e.g. "ToolboxAI") and click Create Key.
  4. Copy the key. Anthropic will only show it to you once, so make sure to copy it before closing the dialog.

-- IMAGE COMING SOON --

4. Review data retention​

By default, Anthropic retains your API inputs and outputs for 30 days for safety and security purposes — worth reviewing before connecting ToolboxAI. There's no self-serve toggle to turn this off; it requires contacting Anthropic directly.

  1. In the Anthropic Console, go to Organization Settings → Privacy controls.
  2. Under Data retention, you'll see your organization's default (On · 30 days). If your organization needs Zero Data Retention — for example, HIPAA or other compliance requirements — click Contact support to request it.

-- IMAGE COMING SOON --

Don't confuse this with the Development Partner Program section further down the same page — that only covers sharing Claude Code sessions, not the API traffic ToolboxAI sends, so it doesn't need any changes for ToolboxAI's purposes.

5. Add the key to ToolboxAI​

  1. In Salesforce, go to App Launcher → ToolboxAI.
  2. Click the Settings tab.
  3. Under AI Connections, paste your key into the Claude field.
  4. Click Save.

-- IMAGE COMING SOON --

Once saved, any tool set to a Claude model in Tool Settings will use this key automatically.

Anthropic's security and compliance documentation​

Anthropic maintains SOC 2 Type II, ISO 27001, and ISO 42001 certification, and publishes an annual third-party penetration testing summary. If your organization needs this documentation for your own vendor security review, you can request access directly through Anthropic's Trust Center — no relationship with ToolboxAI is required, since this is Anthropic's own account and infrastructure, not something ToolboxAI operates on your behalf.

Anthropic API reference​

ToolboxAI calls Anthropic's Messages endpoint (/v1/messages) directly with your API key. Full request/response schemas, authentication headers, and sample requests are documented in Anthropic's official API reference.