MCP Tool Reference¶

The Blueprint MCP server provides 39 tools and 3 resources for AI assistants and automated workflows. Use get_tool_cheatsheet() as your entry point.

Overview by category¶

Search & Answers¶

Templates & Phase Content¶

Analysis & Decision-making¶

Terminology & Utilities¶

Guided Workflows¶

Project Setup (3 steps)¶

Gate Review (2 steps)¶

Compliance (2 steps)¶

Sessions & Project Tracking¶

Lead Management (GDPR-compliant)¶

Reminders¶

Resources (read-only)¶

Tool details¶

answer_question¶

Answers a substantive question via semantic search (RAG) followed by keyword fallback.

answer_question(question: str, output_format: str = "markdown") -> str

Returns up to 3 results: the best match with full content, the rest with summaries.

search_content¶

Search documentation by keyword with optional filters.

search_content(query: str, type: str | None = None, phase: int | None = None, layer: int | None = None, tag: str | None = None, output_format: str = "markdown") -> str

lookup_terminology¶

Look up a term in the Blueprint glossary.

lookup_terminology(term: str, output_format: str = "markdown") -> str

get_template¶

Retrieve a template by name (exact or partial match).

get_template(name: str, output_format: str = "markdown") -> str

check_gate_readiness¶

Compare provided evidence against the required gate criteria.

check_gate_readiness(gate: int, evidence: list[str], output_format: str = "markdown") -> str

classify_risk¶

Classify an AI system into an EU AI Act risk tier.

classify_risk(system_description: str, output_format: str = "markdown") -> str

Returns: unacceptable / high / limited / minimal with obligations per tier.

compliance_checklist¶

Generate an article-referenced EU AI Act checklist.

compliance_checklist(description: str, risk_category: str, output_format: str = "markdown") -> str

session_start¶

Start a session to track progress, artifacts, and gate results.

session_start(project_id: str, project_type: str, language: str = "en", output_format: str = "markdown") -> str

Returns a session_id for use in subsequent calls.

get_tool_cheatsheet¶

Returns a table of all tools, when to use them, and what the next step is.

get_tool_cheatsheet(intent: str = "", output_format: str = "markdown") -> str

submit_lead¶

Submit a new lead intake record with GDPR-compliant consent.

submit_lead(name: str, email: str, company: str, role: str, project_phase: str, help_request: str, consent_given: bool, policy_version: str = "1.0", output_format: str = "markdown") -> str

delete_lead_data¶

Erase all lead data for a given email (GDPR right to erasure).

delete_lead_data(email: str, output_format: str = "markdown") -> str

export_lead_data¶

Export all lead data for a given email (GDPR right of access).

export_lead_data(email: str, output_format: str = "markdown") -> str

schedule_reminder¶

Schedule a reminder for a session.

schedule_reminder(session_id: str, due_date: str, message: str, output_format: str = "markdown") -> str

list_reminders¶

List all reminders for a session.

list_reminders(session_id: str, output_format: str = "markdown") -> str

cancel_reminder¶

Cancel a pending reminder.

cancel_reminder(reminder_id: int, output_format: str = "markdown") -> str

list_sources¶

List all authoritative sources used in the Blueprint, with optional topic filter.

list_sources(topic: str = "", output_format: str = "markdown") -> str

Installation¶

Claude Code (CLI)¶

claude mcp add blueprint --transport http https://ai-delivery.io/mcp

Claude Desktop¶

{
  "mcpServers": {
    "blueprint": {
      "type": "http",
      "url": "https://ai-delivery.io/mcp"
    }
  }
}

Cursor / other MCP clients¶

Add an HTTP MCP server with URL https://ai-delivery.io/mcp.