New App

v1.0.0

Use when creating a new internal application from scratch, standardizing a 0-to-1 app build workflow, or helping teammates bootstrap a Next.js/FastAPI/Keyclo...

0· 130·0 current·0 all-time
by橙子瓣@lyc-chengzi

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for lyc-chengzi/new-app.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "New App" (lyc-chengzi/new-app) from ClawHub.
Skill page: https://clawhub.ai/lyc-chengzi/new-app
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install new-app

ClawHub CLI

Package manager switcher

npx clawhub@latest install new-app
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
Name, description and SKILL.md all describe a documentation‑first internal app bootstrap workflow (Next.js/FastAPI/Keycloak/Postgres style). There are no unexpected environment variables, binaries, or config paths requested that would be disproportionate to that purpose.
Instruction Scope
The runtime instructions are repository- and process-focused: requirements discovery, docs updates, templates under templates/, and standard delivery checklists. They do not instruct the agent to read unrelated system files, access network endpoints, exfiltrate data, or prompt for unrelated credentials. The skill does instruct the agent to read repository docs (docs/) and provided templates, which is appropriate for this purpose.
Install Mechanism
No install spec and no code files — instruction-only — so nothing is downloaded or written to disk by an installer.
Credentials
The skill declares no required environment variables, credentials, or config paths. The content explicitly warns against hardcoding secrets and focuses on documenting runtime configuration, which is proportionate.
Persistence & Privilege
always is false and there is no request to modify other skills or system-wide agent settings. disable-model-invocation is false (agent may invoke it autonomously), which is the platform default and acceptable for a user-invocable, workflow-oriented skill.
Assessment
This skill is instruction-only and appears coherent with its stated purpose. Before installing: (1) confirm you trust the skill author and repository contents; (2) review the templates/docs in the repo to ensure they contain no sensitive data or organization-specific secrets; (3) avoid providing any credentials to the skill and do not rely on it to perform automated network operations unless you verify such behaviors; (4) treat any generated implementation suggestions or code as draft — review and lint/test before merging; (5) if you do not want the agent to call this skill autonomously, disable autonomous invocation in your agent configuration or only invoke it manually.

Like a lobster shell, security has layers — review code before you run it.

latestvk97a6gwsrqk590wmztfwsw44cd83jhg7
130downloads
0stars
1versions
Updated 1mo ago
v1.0.0
MIT-0

Internal App Bootstrap

Purpose

Use this skill when a teammate wants to build a new internal application with a consistent, reviewable, documentation-first workflow instead of jumping straight into code.

This skill captures the workflow used to build i-skills from an empty workspace into a working full-stack application with:

  • requirements discovery before implementation
  • documented architecture and workflow decisions
  • a monorepo-style frontend and backend split
  • authentication planning
  • database and schema planning
  • implementation sequencing
  • validation and operational readiness

When To Use

Use this skill when the user asks for any of the following:

  • build a new app from scratch
  • create a standardized internal tool
  • bootstrap a full-stack product with a clear process
  • help a team follow a repeatable app delivery workflow
  • turn an idea into an MVP with aligned frontend, backend, and docs

Do not use this skill for narrow bug fixes, isolated UI tweaks, or one-off backend endpoint changes.

Core Principles

  1. Requirements first
  2. Architecture before implementation
  3. Documentation before code
  4. English and Chinese documentation stay aligned
  5. Small, verifiable milestones
  6. Validation is part of delivery, not an afterthought

Reusable Standards To Apply

When using this skill, treat the repository standards under docs/ as reusable baseline rules, not project-specific trivia.

Documentation authority by domain

  • docs/api-development.md and docs/api-development.cn.md define backend contract, route, schema, migration, authentication, and operational diagnostics expectations.
  • docs/ux-development.md and docs/ux-development.cn.md define navigation, form, feedback, authentication UX, responsive behavior, and content rules.
  • docs/coding-standards.md and docs/coding-standards.cn.md define repository-wide structure, tooling, naming, validation, and change-discipline rules.
  • docs/copilot-guide.md and docs/copilot-guide.cn.md define the required docs-first workflow and the order in which work should happen.

Common rules worth reusing in new app bootstraps

  • Keep frontend, backend, and docs aligned around the same product vocabulary.
  • Keep public contracts explicit and stable.
  • Put business rules in backend services, not only in frontend behavior.
  • Keep route handlers and page components thin.
  • Use one JavaScript package manager consistently.
  • Treat validation, health checks, and diagnostics as part of delivery.
  • Prefer explicit auth, storage, and database boundaries over hidden side effects.
  • If production uses PostgreSQL, use explicit schema management and migrations.
  • If UI depends on backend data or permissions, design both fallback and error states intentionally.
  • Do not call work complete if paired documentation and validation are missing.

Standard Workflow

1. Clarify the business problem

Start by extracting the operational shape of the app.

Capture at minimum:

  • who will use it
  • what the primary jobs-to-be-done are
  • whether the app is internal or external
  • whether data is sensitive or regulated
  • what the MVP must include
  • what is explicitly out of scope

Use the requirements template in templates/requirements-template.md and templates/requirements-template.cn.md.

2. Identify non-negotiable constraints

Before proposing architecture, confirm constraints such as:

  • authentication provider
  • deployment environment
  • database target
  • storage strategy
  • whether uploads are required
  • access control and visibility rules
  • whether the app executes user-provided artifacts or only stores/references them
  • mobile adaptation expectations

3. Propose a concrete solution

After the requirements are clear, propose a solution that covers:

  • frontend framework
  • backend framework
  • database
  • auth integration
  • storage approach
  • recommended project structure
  • MVP page map or workflow map
  • major risks and tradeoffs

Do not start implementation until the solution is confirmed.

4. Create or update documentation first

Before code changes, write or update the relevant English and Chinese docs.

Always read the matching standards in docs/ first and treat them as the authority for what needs to be documented and validated.

For a new product or workflow, document at least:

  • API expectations
  • UX expectations
  • coding standards if new conventions are introduced
  • workflow guidance if team process is affected

5. Establish the workspace and toolchain

Set up the project structure intentionally.

Typical baseline:

  • frontend under apps/web
  • backend under apps/api
  • docs under docs
  • shared workflow guidance under .github

Normalize the toolchain early:

  • choose one JavaScript package manager
  • define dev tasks
  • define environment file conventions
  • avoid mixed lockfiles or partial scaffolding state

6. Build the MVP in thin vertical slices

Prefer end-to-end slices over isolated piles of code.

A good slice order is:

  1. app shell and navigation
  2. backend models and core routes
  3. frontend pages wired to documented contracts
  4. upload or creation workflow
  5. review or approval workflow
  6. authentication integration
  7. operational tooling

7. Integrate auth and identity deliberately

If authentication is required:

  • separate browser and server auth concerns
  • avoid hardcoding secrets
  • document runtime configuration
  • expose a minimal profile or debug view for identity inspection when needed
  • derive sensitive identity fields on the server

8. Treat the database as a product concern

If production uses PostgreSQL:

  • configure the target database and schema explicitly
  • avoid relying on implicit defaults
  • manage schema changes through Alembic migrations
  • provide diagnostics for connectivity, schema, and revision state

9. Add operational affordances

Before calling the app usable, add:

  • dev tasks for common operations
  • health or diagnostics commands
  • migration commands
  • clear local startup instructions

10. Validate before handoff

Validation should include the touched surfaces.

Typical checks:

  • frontend build
  • backend import or startup sanity check
  • representative endpoint checks
  • auth flow validation when applicable
  • migration or schema validation when applicable

Use the delivery checklist in templates/delivery-checklist.md and templates/delivery-checklist.cn.md.

Expected Deliverables

A complete bootstrap task should usually produce:

  • a documented requirements summary
  • a confirmed architecture proposal
  • updated English and Chinese docs
  • runnable frontend and backend foundations
  • clear environment and task configuration
  • validated core flows
  • a concise change summary and next steps

Output Pattern

When using this skill, structure the work like this:

  1. Requirements summary
  2. Proposed architecture and MVP boundaries
  3. Documentation updates
  4. Implementation plan
  5. Code and configuration changes
  6. Validation results
  7. Follow-up recommendations

Common Failure Modes

Avoid these mistakes:

  • coding before confirming requirements
  • skipping the relevant docs/ files and inferring rules only from existing code
  • letting docs lag behind implementation
  • using mock behavior as the contract
  • mixing package managers
  • wiring auth without a debug path
  • relying on automatic database mutation in shared environments
  • stopping at scaffolding without validating real workflows

Skill Assets

  • templates/requirements-template.md
  • templates/requirements-template.cn.md
  • templates/delivery-checklist.md
  • templates/delivery-checklist.cn.md

Use the templates directly when preparing a new app kickoff or reviewing whether a build is truly complete.

Comments

Loading comments...