MCP Integration
DAC exposes a Model Context Protocol (MCP) server that lets you query your product maturity data, run scoring, and get AI-driven roadmap recommendations directly from any MCP-compatible AI assistant, including Claude, Cursor, and others. Instead of switching between your AI assistant and the Dacard.ai app, you can ask questions like “what are my top 3 priorities this week?” or “show me the transition plan from AI-Curious to AI-Enhanced” and get structured, framework-grounded answers inline.Access tiers
| Plan | MCP access |
|---|---|
| Free | Read-only (framework data and scoring engine only) |
| Starter+ | Full access (all 8 tools, including score_product and roadmap generation) |
Connecting
Get your API key
Go to Settings > API in the Dacard.ai app and generate an API key. Starter plan or above required for full access.
Add to your AI assistant
Add the Dacard MCP server to your assistant’s MCP configuration. The server endpoint and setup instructions are shown in Settings > API.
MCP runs over HTTP. Your AI assistant needs to support MCP tool calls (Claude via Claude.ai or API, Cursor, Continue, and others).
Available tools
get_framework_overview
get_framework_overview
Returns the full structure of the AI-Native Product Team Framework: 5 maturity stages, 24 dimensions, scoring engine details, and cluster definitions.Use for: Understanding the framework before running scoring, or asking your assistant to explain the methodology.Parameters: None.
get_stage_details
get_stage_details
Returns the deep-dive for a specific maturity stage: signals (team, tooling, outcomes), how each of the 24 dimensions manifests at that stage, anti-patterns to avoid, and transition triggers to reach the next stage.Use for: “What does AI-First look like in practice?” or “What anti-patterns should I avoid at AI-Curious?”Parameters:
stage, one of legacy, ai-curious, ai-enhanced, ai-first, ai-native.get_lifecycle_tasks
get_lifecycle_tasks
Returns tasks from the AI-Native Product Development Lifecycle: 6 stages (Specify, Context, Orchestrate, Validate, Ship, Compound) replacing the traditional SDLC.Use for: “What tasks should my team be doing in the Orchestrate stage?” or pulling all lifecycle tasks for a full view.Parameters:
stage (optional), filter to a specific lifecycle stage. Omit to return all stages.score_product
score_product
Given answers to all 24 dimension questions (each scored 1-4), calculates the AI-native maturity score (24-96), determines the tier, and provides recommendations. Accepts optional company context for richer analysis.Use for: Running a scoring session through your AI assistant without opening the app.Parameters: 24 dimension scores (1-4 each), optional
context (company stage, team size, product category).get_dimension_progression
get_dimension_progression
Shows how a specific dimension evolves across all 5 maturity stages, including the inflection point where the biggest capability jump happens.Use for: “How does decision_quality evolve from Legacy to AI-Native?” or planning a specific capability improvement.Parameters:
dimension, any of the 24 dimension IDs (e.g., decision_quality, feedback_loop_quality).get_personalized_roadmap
get_personalized_roadmap
Given 24 dimension scores and optional company context, generates a prioritized AI transformation roadmap: cluster analysis, monthly action plan, weak-area diagnostics, and tool recommendations.Use for: “Build me a 3-month roadmap based on my scores” or pulling the same roadmap the app generates.Parameters: 24 dimension scores (1-4 each), optional
context.get_action_items
get_action_items
Returns the top 3-5 prioritized this-week actions based on 24 dimension scores. Surfaced from the highest-impact, most immediately actionable tasks in the personalized roadmap.Use for: Quick “what should I do next?” queries without needing the full roadmap.Parameters: 24 dimension scores (1-4 each), optional
context.get_transition_plan
get_transition_plan
Returns a detailed transition plan between two maturity stages: per-dimension steps needed, anti-patterns to avoid, and transition triggers to watch for. Accepts optional current scores for personalized emphasis.Use for: “How do we move from AI-Curious to AI-Enhanced?” or strategic planning workshops.Parameters:
current_stage, target_stage, optional scores.Example prompts
Once connected, try these with your AI assistant:Framework overview
“Use Dacard to explain the 5 maturity stages and what separates AI-First from AI-Native.”
Stage deep-dive
“Use Dacard to show me all the signals and anti-patterns at the AI-Curious stage.”
Quick actions
“Based on my scores [paste scores], what are my top 3 priorities this week?”
Transition plan
“Use Dacard to build a transition plan from AI-Enhanced to AI-First for a 12-person team.”
Read-only vs. full access
Free plan MCP access coversget_framework_overview, get_stage_details, get_lifecycle_tasks, and get_dimension_progression, the framework reference tools. Scoring and roadmap tools (score_product, get_personalized_roadmap, get_action_items, get_transition_plan) require Starter plan or above.