brim/break
Back to stack
02 · Layer

Claude.

Outbound, ICP match, account briefings — as prompt-as-config.

Claude wired into HubSpot records via the Model Context Protocol. No generic GPT mumbling: custom prompts per use case, batch-ready, token logs in BigQuery.

1M+
Tokens / day
<2min
Account brief
~€0.05
Cost / enrichment
100%
Auditable
We've all been there

What you've probably seen.

  • ICP match by gut feel in the sales meeting
  • Outbound personalization manual per lead
  • Account briefings clicked together in 5 browser tabs before calls
  • No audit trail of who personalized what when
How we solve it

How we set it up.

  • Prompts versioned in Git, deployed via CI
  • Batch enrichment over the entire CRM list
  • MCP server feeds live records to the model
  • Token cost transparent per workflow + per lead
Toolchain
Anthropic APIMCP ServerVercel AI SDKPrompts-as-CodeTool CallsStreamingPrompt CachingToken LogsAnthropic APIMCP ServerVercel AI SDKPrompts-as-CodeTool CallsStreamingPrompt CachingToken Logs
Example workflow

Example: ICP-match enrichment

  1. 01Account ID arrives from HubSpot
  2. 02MCP fetches properties + LinkedIn snapshot
  3. 03Claude analyzes and scores ICP fit
  4. 04Score + brief land back in HubSpot

Want us to build this for you?.

30 min demo. We walk you through a real setup, live.