sdd-research

Synthesize parallel codebase and external research into a research.md document.

1|Updated Apr 25, 2026
One-click install
npx skills add https://github.com/dariomazzitellireplik-coder/sdd-skills --skill sdd-research-dariomazzitellireplik-coder
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sdd-research
Source: https://github.com/dariomazzitellireplik-coder/sdd-skills/tree/main/skills/sdd-research
Command: npx skills add https://github.com/dariomazzitellireplik-coder/sdd-skills --skill sdd-research-dariomazzitellireplik-coder

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill addresses the challenge of conducting thorough research during the second phase of Spec-Driven Development (SDD), ensuring comprehensive understanding and exploration of codebase, external sources, and technical options before planning.

Core Features & Use Cases

  • Parallel Subagent Research: Initiates parallel research using subagents for codebase exploration, analysis, and architectural review.
  • Synthesis of Findings: Aggregates and synthesizes findings into a research.md document for further planning.
  • Use Case: After a requirement is approved, use this Skill to research potential solutions, codebase patterns, and architectural options before planning the execution.

Quick Start

Run '/sdd-research' after the requirement is approved to initiate the research phase.

Frequently Asked Questions about sdd-research

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I conduct codebase analysis and architectural review before planning new features?

Codebase analysis and architectural review are conducted by invoking parallel subagents to explore patterns and synthesize findings into a research document for planning. This ensures comprehensive understanding of technical options before execution.

What is the best way to research technical options and codebase patterns in parallel?

Researching technical options in parallel is best handled by deploying subagents simultaneously for codebase exploration and external source analysis. Findings are then aggregated and synthesized into a single document for downstream planning.

When do I need to run research during spec-driven development?

Research during spec-driven development is needed immediately after a requirement is approved and before the planning phase begins. It explores potential solutions, codebase patterns, and architectural options to inform execution.

Can I use parallel processing for code analysis without manual subagent management?

Parallel processing for code analysis can be used without manual management by running the research command, which automatically initiates parallel subagents. These subagents independently handle codebase exploration and external source analysis.

Does spec-driven development research work with external sources and technical option exploration?

Spec-driven development research works with both external sources and technical option exploration by dispatching parallel subagents for each domain. The subagents analyze the codebase and external references, then synthesize all findings together.

What are the limitations of synthesizing research findings into a single document?

Synthesizing research findings into a single document requires parallel processing and synthesis capabilities to function correctly. Without these capabilities, the aggregation of codebase and external source analysis cannot be properly prepared for planning.