MAP (Mediating Assessments Protocol) Skill

This skill implements Daniel Kahneman's MAP to reduce bias and noise by forcing independent evaluation of key dimensions before a final holistic judgment.

Philosophy

To prevent the "Halo Effect" and "WYSIATI" (What You See Is All There Is), complex decisions must be factored into independent sub-questions.

Workflow

1. •

   Decompose the Decision: Identify the core attributes of the decision. For standard corporate audits, use the MAP Trio:

   • •Evidence Quality: How robust, verified, and complete is the data? (1-10)
   • •Market Stability: How volatile are the external conditions? (1-10)
   • •Team Capability: Does the executing team have a track record of success? (1-10)
2. •

   Independent Assessment (Simulated):

   • •Launch THREE (3) task sub-agents (judges) to rate these dimensions independently.
   • •Constraint: Judges must rate each dimension without seeing the ratings of others.
3. •

   Generate Decision Scorecard:

   • •Compile the ratings into a structured artifact.
   • •Calculate Level Noise (Variance) for each dimension.
4. •

   Output Artifact (Decision Scorecard): Produce a JSON/Markdown table highlighting agreement vs. divergence.

   json

   {
     "artifact_type": "decision_scorecard",
     "dimensions": {
       "evidence_quality": { "scores": [7, 8, 4], "variance": 1.7, "status": "High Noise" },
       "market_stability": { "scores": [9, 9, 8], "variance": 0.47, "status": "Consensus" },
       "team_capability":  { "scores": [6, 5, 6], "variance": 0.47, "status": "Consensus" }
     },
     "holistic_recommendation": "Pause for evidence verification."
   }
   

Usage

Trigger this skill when the user asks for a "Scorecard", "MAP assessment", or "Structured Evaluation" of a proposal.