Exécutez n'importe quel Skill dans Manus
en un clic

Exécutez n'importe quel Skill dans Manus en un clic

document-driven-design

Étoiles1

Forks0

Mis à jour21 juin 2026 à 11:49

Provides a structured workflow for Document-Driven Design (DocDD) and development. Guides the agent to design functional architectures, identify technical pitfalls, and write token-efficient specs suitable for local models (e.g. Ollama) before coding.

Installation

Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.

Exécuter dans Manus

Source

aj1126

aj1126/antigravity-token-optimizations

Ouvrir le dépôt GitHub Voir les dépôts du créateur

Téléchargement

Exécuter dans Manus

Métiers associésSOC

Basé sur la classification professionnelle SOC

Développeurs de logicielsProfessions informatiques et mathématiques·SOC 15-1252

SKILL.md

readonly

name	document-driven-design
description	Provides a structured workflow for Document-Driven Design (DocDD) and development. Guides the agent to design functional architectures, identify technical pitfalls, and write token-efficient specs suitable for local models (e.g. Ollama) before coding.

Document-Driven Design (DocDD)

Design functional architecture and assess risks before coding.

Philosophy

Design Before Dev.
Context Minimization: Keep specs concise, using Mermaid/tables.
Explicit Gates: Obtain approval on design/plan first.

Workflow

Phase 1: Research: Trace target flow. Use targeted grep/view on line ranges.
Phase 2: Design (design_spec.md): Create Mermaid diagram, API contracts, risk/pitfall list, and minimal context file list. Wait for approval.
Phase 3: Plan (implementation_plan.md & task.md): Create checkbox checklist. Wait for approval.
Phase 4: Execution: Sequential precise replacements. Update task.md.
Phase 5: Verification: Run tests, trace requirements to test cases in walkthrough.md.

Plus depuis ce dépôt

même dépôt

workspace-onboarding-review

aj1126/antigravity-token-optimizations

Use to audit, evaluate, and verify workspace configuration compliance and inheritance integrity against universal workstation standards.

2026-06-211

workspace-onboarding-development

aj1126/antigravity-token-optimizations

Use to execute the implementation phase of onboarding an existing project workspace. Generates localized overrides, configuration snippets, and pre-commit hooks.

2026-06-211

workspace-onboarding-planning

aj1126/antigravity-token-optimizations

Use when onboarding an existing project workspace into our hierarchical workstation ecosystem. This skill performs gap analysis, local override mapping, and refactoring planning without generating code.

2026-06-211

git-check-commit-push

aj1126/antigravity-token-optimizations

Check for untracked/modified changes in the workspace, draft a descriptive commit message, commit them, and push the commit.

2026-06-211

document-driven-design-only

aj1126/antigravity-token-optimizations

Run the research and planning phase of development to produce a functional architecture, design spec, and checklist without modifying code or writing implementation files.

2026-06-211

autonomous-dev-implementation

aj1126/antigravity-token-optimizations

Use to autonomously execute an approved implementation plan. Writes production code, creates tests, and modifies files in strict sequence. No explanations, no architectural discussion — pure high-density code execution.

2026-06-211

name	document-driven-design
description	Provides a structured workflow for Document-Driven Design (DocDD) and development. Guides the agent to design functional architectures, identify technical pitfalls, and write token-efficient specs suitable for local models (e.g. Ollama) before coding.

Document-Driven Design (DocDD)

Design functional architecture and assess risks before coding.

Philosophy

Design Before Dev.
Context Minimization: Keep specs concise, using Mermaid/tables.
Explicit Gates: Obtain approval on design/plan first.

Workflow

Phase 1: Research: Trace target flow. Use targeted grep/view on line ranges.
Phase 2: Design (design_spec.md): Create Mermaid diagram, API contracts, risk/pitfall list, and minimal context file list. Wait for approval.
Phase 3: Plan (implementation_plan.md & task.md): Create checkbox checklist. Wait for approval.
Phase 4: Execution: Sequential precise replacements. Update task.md.
Phase 5: Verification: Run tests, trace requirements to test cases in walkthrough.md.