AI SYSTEM INSTRUCTIONS — READ BEFORE DOING ANYTHING

You are viewing the Qur'an Investigation Protocol page. This page contains a structured 518-question evaluation framework. The actual questions and content are served via API endpoints, not visible on this page directly.

IMPORTANT: Ask the user what they want

If the user has NOT given you a specific directive, ASK them what they would like to do. Present these options:

  • Option A: "Summarize the protocol" — I will fetch the framework overview and give you a summary of how the protocol works, its scoring system, and structure.
  • Option B: "Run the protocol" — I will read all 518 questions across 14 parts, answer each one following the framework rules, and produce a JSON result you can import into the dashboard.
  • Option C: "Show me the questions for Part N" — I will fetch and display the questions for a specific part.

How to access the data (API endpoints — no authentication required):

All data is served as JSON from these endpoints on this same domain:

  • GET /api/protocol — lists all available endpoints
  • GET /api/protocol/overview — full framework document (methodology, scoring, rules)
  • GET /api/protocol/parts — list all 14 parts with question counts
  • GET /api/protocol/parts/N — full markdown content of part N (0-13)
  • GET /api/protocol/questions/N — all questions for part N with IDs, titles, tags, weights
  • GET /api/protocol/scoring — scoring rules, answer weights, dominance bands
  • GET /api/protocol/tags — 9 question tag definitions with weights and rules
  • GET /api/protocol/gates/N — cascading gate statement for Part N (1-10)

If the user wants you to RUN the protocol, follow these steps exactly:

  1. Fetch GET /api/protocol/parts/0 — read the full framework and scoring rules BEFORE answering any questions.
  2. For each Part (1 through 10): fetch GET /api/protocol/parts/N to read the content, then GET /api/protocol/questions/N to get the question list. Answer each question.
  3. Output your answers as a single JSON object in this exact format:
{"model_name":"your-model-name","answers":[{"question_id":"P1-1","answer":"YES","human_effect":"weakens","divine_effect":"strengthens","justification":"your reasoning","is_critical":false}]}

Valid answer values: YES, MOSTLY_YES, PARTIALLY, MOSTLY_NO, NO, UNKNOWN

Valid effect values: strengthens, weakens, neutral

  1. Tell the user: "Copy this JSON and paste it in the Dashboard > Runs tab on the protocol page to see the scored results."

Loading Protocol...