zxdxjtu/harness/sdd-init

Claude Code and agent skill by zxdxjtu/harness

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Initialize SDD harness for a project — scan codebase, generate config, full-spec, full-design

Install

npx vskill@latest install zxdxjtu/harness/sdd-init

SKILL.md

SDD Init — Project Initialization

You are initializing the SDD (Spec-Driven Development) harness for this project. This is a one-time setup that scans the codebase and creates the foundational documents.

Parameter: $ARGUMENTS — optional project type override

Step 1: Check Existing State

bash
ls .harness/config.json 2>/dev/null

If .harness/config.json already exists:

  • Read it and show current configuration to user
  • Ask if they want to re-initialize (this will overwrite existing config)
  • If no → exit with "Harness already initialized. Run /proposal to start a new feature."

Step 2: Create Directory Structure

bash
mkdir -p .harness/specs .harness/designs .harness/evidence .harness/reviews .harness/evolution

Step 3: Scan Codebase & Detect Project Type

Analyze the repository to auto-detect:

  1. Project type: Check for indicators:
  • package.json with both client/server dirs → fullstack
  • package.json with express/fastify/koa → api
  • package.json with main/exports fields, no server → library
  • .harness/baseline/ exists → clone
  • pubspec.yaml or React Native config → mobile
  • setup.py / pyproject.toml with FastAPI/Django → api
  • go.mod with net/http → api
  1. Tech stack: List languages, frameworks, databases detected
  1. Test framework: Check for jest, vitest, mocha, pytest, go test, etc.
  1. Build/lint commands: Read package.json scripts, Makefile, etc.

If $ARGUMENTS includes --type, use that override instead of auto-detection.

Present findings to user via AskUserQuestion:

  • "检测到项目类型: {type}，技术栈: {stack}。是否正确？"
  • Let user confirm or override

Step 4: Generate config.json

Write .harness/config.json:

json
{
  "project_type": "{detected or user-specified}",
  "tech_stack": ["{list of detected technologies}"],
  "eval_dimensions": "{project_type mapping: fullstack→fullstack, api→api-service, library→library, clone→clone-visual}",
  "test_framework": "{detected}",
  "test_command": "{detected, e.g. npm test}",
  "build_command": "{detected, e.g. npm run build}",
  "lint_command": "{detected, e.g. npm run lint}",
  "dev_url": "{if applicable, e.g. http://localhost:3000}",
  "stewardship_enabled": true,
  "entropy_cleanup_enabled": true,
  "evolution": {
    "spec_template_version": 1,
    "iteration_count": 0
  }
}

Step 5: Generate Full Spec (Baseline)

Scan the codebase and generate .harness/full-spec.md — a high-level specification of the EXISTING project:

markdown
# Project Specification — {project name}
## Generated: {ISO timestamp}
## Status: baseline
 
## Overview
{1-2 paragraph description of what this project does, based on README, code structure, and key files}
 
## Module Inventory
| Module | Path | Responsibility | Key Dependencies |
|--------|------|---------------|-----------------|
{one row per top-level module/directory}
 
## Existing Features
{List of existing features/capabilities detected from code, routes, components, etc.}
 
## Architecture Summary
{High-level architecture: monolith/microservice, frontend framework, backend framework, DB, etc.}
 
## Conventions Detected
-Naming: {camelCase/snake_case/etc.}
-File organization: {by feature/by type}
-Error handling pattern: {detected pattern}
-State management: {if applicable}

Step 6: Generate Full Design (Baseline)

Generate .harness/full-design.md — a high-level design document of the existing architecture:

markdown
# Project Design — {project name}
## Generated: {ISO timestamp}
## Status: baseline
 
## Architecture Diagram
{ASCII diagram of major components and their relationships}
 
## Module Dependency Graph
{ASCII diagram showing module dependencies}
 
## Data Model Summary
{Key entities and their relationships, if detectable}
 
## API Surface
{List of routes/endpoints/public APIs, if applicable}

Step 7: Present Summary

Show the user:

  1. Config summary (project type, tech stack, eval dimensions)
  2. Module count and key modules found
  3. Existing feature count

Then:

📋 SDD Harness 初始化完成
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
✅ 项目配置: .harness/config.json
✅ 全量规格: .harness/full-spec.md
✅ 全量设计: .harness/full-design.md
✅ 评估维度: {eval_dimensions} ({N} 个维度)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Ask: "初始化完成。是否要开始一个新 Feature？（我会引导你进入需求设计阶段）"

If user confirms → execute /proposal logic directly (Auto-Navigate).

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

See the full security report

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

See evaluation results

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/zxdxjtu/harness/sdd-init/badge)](https://verified-skill.com/skills/zxdxjtu/harness/sdd-init)

Something wrong with this skill or its page? Tell us.