sdd-apply

Executes predefined SDD tasks with support for multiple metadata backends.

2|Updated Jul 29, 2025
One-click install
npx skills add https://github.com/kurojs/EnderDots --skill sdd-apply-kurojs
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sdd-apply
Source: https://github.com/kurojs/EnderDots/tree/main/.config/opencode/skills/sdd-apply
Command: npx skills add https://github.com/kurojs/EnderDots --skill sdd-apply-kurojs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps an executor safely implement SDD change tasks by turning orchestrator-assigned specs and designs into correct, reviewable code updates while preserving prior progress.

Core Features & Use Cases

  • SDD executor workflow: Implements one or more assigned tasks using the provided specs and design as acceptance criteria.
  • Artifact-aware persistence: Supports engram, openspec, hybrid, and none modes to read required artifacts and persist apply progress.
  • Workload decision enforcement: Applies the “review forecast” gate to ensure the orchestrator/user provides a resolved delivery path (auto-chain, chained slice, single PR, or size:exception) before writing code.
  • Strict TDD mode support: When enabled, requires a TDD cycle evidence table and follows the strict RED→GREEN→TRIANGULATE→REFACTOR execution contract.
  • Cumulative progress merge: Reads existing apply-progress and merges newly completed tasks without losing prior completions.

Quick Start

Load the sdd-apply skill as an executor, then implement only the specific tasks assigned for the given change name while updating tasks completion and saving apply-progress according to the selected artifact mode.

Frequently Asked Questions about sdd-apply

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

FAQPage Schema
How do I implement SDD change tasks without losing prior apply-progress?

To implement SDD change tasks safely, load an executor skill that reads existing apply-progress and merges newly completed tasks. This preserves prior completions while applying assigned task slices against provided specifications and design decisions.

What is the strict TDD cycle for SDD apply-phase execution?

The strict TDD cycle for SDD apply-phase execution follows the RED, GREEN, TRIANGULATE, and REFACTOR sequence. When strict mode is enabled, executors must maintain a TDD cycle evidence table to verify each step before proceeding.

How do artifact persistence modes affect SDD task implementation?

Artifact persistence modes determine how required task artifacts are read and apply-progress is saved during SDD implementation. Supported modes include engram, openspec, hybrid, and none, each defining different storage and retrieval strategies for task artifacts.

Can I execute multiple assigned task slices in a single PR workflow?

Executing multiple assigned task slices requires passing the review forecast gate to resolve the delivery path. The executor enforces workload and PR boundary constraints, allowing auto-chain, chained slice, single PR, or size exception decisions before writing code.

What happens if delivery decisions are missing before SDD apply-phase execution?

If delivery decisions are missing, the apply-phase execution halts at the review forecast gate. The executor requires a resolved delivery path from the orchestrator or user to ensure workload and PR boundary constraints are met before writing any code.

Does SDD apply-phase execution work without TDD requirements?

SDD apply-phase execution works without strict TDD requirements by implementing assigned tasks normally using specifications as acceptance criteria. Strict TDD mode is optional and only enforces the RED, GREEN, TRIANGULATE, REFACTOR contract when explicitly enabled.