Public MCP

Review buying-signal evidence from your AI client.

Connect an OpenAI or Claude MCP client to https://intakra.com/mcp. The public endpoint evaluates only the evidence you supply and keeps every result inside a strict human-review boundary.

Available tools

score_buying_signal_evidence

Scores caller-supplied fit criteria and business-signal evidence. It returns a review-only fit, timing, evidence-quality, and priority scorecard.

prepare_why_now_review

Prepares fit and why-now guidance that keeps each fact labeled caller-supplied and unverified. It never drafts outreach.

Safety and evidence boundary

  • No Intakra login or API key is required.
  • No workspace, CRM, contact, account, signal, or customer data is read.
  • No source is opened and no model or paid provider is called.
  • No input or result is persisted by the MCP implementation.
  • No outreach is drafted or sent, and no CRM action is available.
  • Every result is a draft for human review, not a buying-intent prediction.

Keep private workspace data in the authenticated /api/mcp integration. Do not submit personal, confidential, or credentialed data to the public endpoint.

Need integration support?

Send the client name, request ID, and non-sensitive error message. Never send API keys or private evidence in a support request.

Contact Intakra support