sdd-apply

Implement code changes from specs, design, and active tasks in an SDD pipeline.

Updated Apr 8, 2026
One-click install
npx skills add https://github.com/aroca89/roca_teams --skill sdd-apply-aroca89
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sdd-apply
Source: https://github.com/aroca89/roca_teams/tree/main/skills/sdd-apply
Command: npx skills add https://github.com/aroca89/roca_teams --skill sdd-apply-aroca89

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Dedicated implementation agents often miss the structure and persistence needed for medium or large feature changes, so this skill ensures the apply phase reads specs, design, and task context before writing any production code.

Core Features & Use Cases

  • Spec-compliant implementation workflow: load proposal, spec, design, and tasks to deliver code that matches requirements and existing patterns.
  • Mode-aware persistence: follow engram, openspec, hybrid, or none conventions to update artifacts, tasks, and progress reports without breaking the pipeline.
  • TDD and reporting discipline: detect TDD rules, execute the RED→GREEN→REFACTOR cycle when required, and produce the structured envelope expected by the orchestrator for verification.

Quick Start

Ask the orchestrator to activate the sdd-apply skill so it can read specs, design, and tasks before implementing code.

Frequently Asked Questions about sdd-apply

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

FAQPage Schema
How do I implement change tasks while maintaining structured progress in an SDD pipeline?

Implement change tasks by using sdd-apply to read specs, design, and active tasks before writing code, then mark tasks.md and persist progress for the orchestrator. It ensures implementation agents follow structured progress tracking throughout the pipeline.

What is the TDD workflow for spec-compliant implementation and how does it report progress?

The TDD workflow detects active TDD rules and executes the RED→GREEN→REFACTOR cycle, then produces a structured progress reporting envelope expected by the orchestrator for verification. This ensures spec-compliant implementation discipline.

How do I update artifacts and tasks without breaking the SDD pipeline persistence conventions?

Update artifacts and tasks by detecting the active mode—engram, openspec, hybrid, or none—and following its specific persistence conventions. This mode-aware persistence prevents breaking the pipeline while maintaining structured progress reports.

Can I use automated task implementation without loading design specs first?

No, automated task implementation requires loading the proposal, spec, design, and tasks context first to deliver code that matches requirements and existing patterns. Skipping this step misses the structure needed for medium or large feature changes.

Why does my implementation agent miss persistence and task updates during feature changes?

Dedicated implementation agents often miss persistence because they lack the discipline to read specs and design before coding. sdd-apply solves this by enforcing mode detection, task updates, persistence calls, and structured progress reporting.

Does sdd-apply work with orchestrator-assigned implementation tasks in hybrid mode?

Yes, sdd-apply supports hybrid mode detection alongside engram, openspec, and none modes. It follows the appropriate conventions to update artifacts, tasks, and progress reports when the orchestrator assigns implementation tasks.