analyze

Workspace Resolution

1. If $ARGUMENTS specifies a project name, use workspaces/$ARGUMENTS/
2. Otherwise, use the most recently modified directory under workspaces/ (excluding instructions/)
3. If no workspace exists, ask the user to create one first
4. Read all files in workspaces/<project>/briefs/ for user context (this is the user's input surface)

Phase Check

• Output goes into workspaces/<project>/01-analysis/, workspaces/<project>/02-plans/, and workspaces/<project>/03-user-flows/

Execution Model

This phase executes under the autonomous execution model (see rules/autonomous-execution.md). All analysis, deliberation, and recommendations MUST assume autonomous AI agent execution — not human team constraints. Do not estimate effort in human-days. Do not constrain recommendations by team size or hiring. Recommend the technically optimal approach; agents scale horizontally.

Workflow

1. Be explicit about objectives and expectations

Understand the product idea before diving into research.

2. Perform Deep Research

Document in detail in workspaces/<project>/01-analysis/01-research.

• Use as many subdirectories and files as required
• Name them sequentially as 01-, 02-, etc, for easy referencing

3. Ensure strong product focus

Keep this soft rule in mind for everything:

• 80% of the codebase/features/efforts can be reused (agnostic)
• 15% of client specific requirements goes into consideration for self-service functionalities that can be reused (agnostic)
• 5% customization

Steps:

1. Research thoroughly and distill value propositions and UNIQUE SELLING POINTS
   • Scrutinize and critique the intent and vision, focusing on perfect product-market fit
   • Research competing products, gaps, painpoints, and any other information that helps build solid value propositions
   • Define unique selling points (not the same as value propositions) — be extremely critical and scrutinize them
2. Evaluate using platform model thinking
   • Seamless direct transactions between users (producers, consumers, partners)
     • Producers: Users who offer/deliver a product or service
     • Consumers: Users who consume a product or service
     • Partners: To facilitate the transaction between producers and consumers
3. Evaluate using the AAA framework
   • Automate: Reduce operational costs
   • Augment: Reduce decision-making costs
   • Amplify: Reduce expertise costs (for scaling)
4. Features must cover network behaviors for strong network effects
   • Accessibility: Easy for users to complete a transaction (activity between producer and consumer, not necessarily monetary)
   • Engagement: Information useful to users for completing a transaction
   • Personalization: Information curated for an intended use
   • Connection: Information sources connected to the platform (one or two-way)
   • Collaboration: Producers and consumers can jointly work seamlessly