Structured-Prompt-Driven Development (SPDD)

2026-04-28 · Source: Martin Fowler · Field: Technology & Digital — Software Development & Engineering, Artificial Intelligence & Machine Learning · Depth: Intermediate, extended

Summary

Structured-Prompt-Driven Development (SPDD), a method developed by Thoughtworks' internal IT, makes LLM-assisted changes governable, reviewable, and reusable for teams. Published on April 28, 2026, SPDD treats prompts as first-class artifacts, version-controlled alongside code, and uses the REASONS Canvas—a seven-part structure for guiding prompt generation (Requirements, Entities, Approach, Structure, Operations, Norms, Safeguards). The SPDD workflow enforces a "fix the prompt first, then update the code" rule, supported by the openspdd command-line tool with commands like /spdd-story, /spdd-analysis, /spdd-reasons-canvas, /spdd-generate, /spdd-prompt-update, and /spdd-sync. An example demonstrates enhancing a billing engine with model-aware and multi-plan pricing, detailing steps from initial requirements to unit test generation. SPDD requires developers to master abstraction-first design, alignment, and iterative review, and is highly recommended for scaled, standardized delivery and high-compliance environments.

Key takeaway

For AI Architects and Software Engineers scaling LLM-assisted development, SPDD offers a structured approach to ensure governable, reviewable, and reusable code. You should adopt its "prompt-first" methodology, using tools like openspdd to formalize requirements and design intent before code generation. This reduces rework, improves traceability, and fosters a shared understanding, transforming individual LLM efficiency into reliable team throughput, especially in logic-heavy or high-compliance domains.

Key insights

SPDD transforms LLM assistance into a scalable, governable organizational capability by treating prompts as version-controlled artifacts.

Principles

• Design abstractions before code generation.
• Align intent and constraints upfront.
• Implement disciplined iterative review loops.

Method

SPDD uses the REASONS Canvas to structure prompts, guiding LLMs from intent to execution. It employs openspdd commands for analysis, prompt generation, code generation, and two-way synchronization between prompts and code.

In practice

• Use openspdd for structured prompt management.
• Apply REASONS Canvas for complex business logic.
• Synchronize code and prompts with /spdd-sync.

Topics

• Structured Prompt-Driven Development
• LLM-Assisted Development
• REASONS Canvas
• Prompt Engineering
• Software Governance
• Code Generation

Code references

• gszhangwei/open-spdd
• gszhangwei/token-billing

Best for: AI Engineer, Software Engineer, AI Architect

Related on AIssential

• AI tools for hearing difficulties — helpful or harmful for language learning? — AI assistive technology presents a dilemma for language learners with hearing difficulties, balancing aid with skill development.
• How Enterprise Marketing Teams Are Actually Using AI Agents in 2026 — Enterprise marketing is shifting from human-assisted AI tools to autonomous AI agents that redesign entire workflows.
• [AINews] Microsoft Build: MAI-Thinking-1 and MAI Family models — Microsoft's MAI models showcase a transparent, full-stack AI strategy emphasizing clean data and local agent execution.
• Are we getting better at explaining things because of our interaction with LLMs? — Crafting clear LLM prompts may enhance human explanatory and communication abilities.
• OpenAI expands Codex with role-specific plugins to build a general-purpose app for non-developers — OpenAI is transforming Codex into a general-purpose, role-specific AI work app, expanding beyond coding for non-developers.
• How to Use Claude Managed Agents? — Anthropic's Claude Managed Agents simplifies AI agent deployment by providing a hosted, stateful, and secure execution environment.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Martin Fowler.