Skip to content

LingJueYa/Commons_Plan_Prompt

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Commons Plan

A rigorous AI system prompt for development solution discussions, inspired by parliamentary decision-making. Transform vague requirements into optimal development solutions through structured collective deliberation and debate.

Use this system to facilitate thoughtful, debate-driven discussions about technical choices, architecture decisions, and implementation strategies.

What It Does

Commons Plan guides AI through a three-stage parliamentary-style process for development solution discussions:

  1. Requirements Understanding - Public hearings and iterative clarification until 100% clarity on development needs
  2. Solution Exploration - Generate diverse development proposals → Adversarial debate → Convergent voting
  3. Documentation - Complete legislative record of the development decision with rationale

Why It's Different

Most AI tools jump straight to development solutions. Commons Plan emulates parliamentary deliberation for technical decision-making:

  • Clarifies first - Never assumes, always confirms
  • Explores broadly - Generates 6-8 fundamentally different approaches
  • Challenges rigorously - Red team critiques every solution
  • Decides systematically - Multi-criteria weighted scoring
  • Documents completely - Full legislative record (ADR format)

Core Methodology

Built on research-backed frameworks:

  • Chain-of-Thought Reasoning - Step-by-step logical progression
  • Adversarial Collaboration - Red team / Blue team critique
  • Pre-Mortem Analysis - Identify failures before they happen
  • Multi-Criteria Decision Analysis - Weighted scoring across dimensions
  • Architecture Decision Records - Complete traceability

Process Flow

User Input 
    ↓
Clarification Loop (2-3 rounds)
    ↓
Divergent Generation (6-8 diverse solutions)
    ↓
Adversarial Debate (Red team vs Blue team)
    ↓
Convergent Synthesis (MCDA scoring → Selection)
    ↓
Final Recommendation + ADR Documentation

Use Cases

Perfect for development solution discussions:

  • Technology choices - Which framework/library to adopt?
  • Architecture decisions - Which system design pattern?
  • Implementation strategies - Which approach to build the feature?
  • Development trade-offs - Performance vs maintainability vs speed
  • Team debates - Structured discussion of competing proposals

Example Session: Development Solution Discussion

User: "I need to improve my app's performance"

AI (Parliamentary-style deliberation):

I understand your requirement as:
**Problem**: App performance is suboptimal
**Goal**: Improve performance
**Context**: [Needs clarification]

To ensure clarity, could you clarify:
1. What specific performance issue? (load time/responsiveness/crashes)
2. Current metrics and target metrics?
3. Constraints? (budget/time/can't rewrite)

[After clarification...]

AI generates:

  • 6-8 diverse proposals (caching, CDN, code optimization, architecture change...)
  • Red team debates and critiques each proposal
  • Scores all proposals across effectiveness/feasibility/risk/ROI
  • Recommends optimal solution with full rationale
  • Documents entire deliberation process in legislative record format (ADR)

Key Features

Never Skips Understanding

Enforces confirmation gate before solution generation.

True Diversity

Solutions span: Conservative | Innovative | Minimal | Scalable | Cost-Optimized | Speed-Optimized | Quality-Optimized | Hybrid

Adversarial Rigor

Every solution faces:

  • Fatal flaw identification
  • Assumption challenges
  • Pre-mortem failure scenarios
  • Hidden complexity exposure

Evidence-Based Selection

Multi-criteria scoring with:

  • Customizable weights based on priorities
  • Transparent rationale for every score
  • Synthesis check for hybrid solutions

Complete Traceability

Generates ADR documenting:

  • All options explored
  • Why each was eliminated/retained
  • Trade-offs accepted
  • Decision rationale
  • Review schedule

Quality Guarantees

The prompt includes self-checks:

  • Never skip Stage 1 confirmation
  • No solutions with unaddressed fatal flaws
  • No arbitrary scoring without evidence
  • Always show reasoning transparently
  • Generate both recommendation AND documentation

Optimization

Designed for:

  • Clarity - Simple language, structured format
  • Rigor - No shortcuts, complete analysis
  • Efficiency - Streamlined process (5-7 rounds)
  • Transparency - All reasoning visible
  • Actionability - Immediate next steps included

Credits

Built on research-backed methods:

About

How can we help AI find the best solution?

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published