beifeng724/personal-workbench-skills / skills/personal-workbench-builder
personal-workbench-builder
VerifiedBuild a durable personal or professional workbench as a runnable application, or produce a six-stage generic prompt pack for building it. Use when the user wants to turn a recurring workflow into a dashboard, management system, workspace, personal operating system, or long-lived website with multiple connected modules and persistent state. Trigger for project, CRM, creator, learning, family, finance, travel, teaching, legal, or team workbenches. Do not trigger for marketing sites, one-off scripts, ordinary coding tasks, or informational questions.
references/verification-contract.md
View on GitHub# Verification Contract
Read this file before implementation planning and again before claiming completion.
## Control inventory
Inventory every visible or reachable control, including navigation, buttons, icon buttons, menus, tabs, filters, forms, editors, uploads, import/export, theme controls, dialogs, and recovery actions.
Record this table in `workbench-spec.md`:
| Control ID | Page/module | Label or accessible name | Intended behavior | Acceptance condition | Test method | Actual result | Evidence | Status |
|---|---|---|---|---|---|---|---|---|
Every implemented control needs an acceptance condition before it can be marked complete. Remove dead controls instead of labeling them "coming soon" unless a confirmed requirement explicitly calls for a disabled placeholder.
## Evidence standard
The following are preconditions, not proof that a feature works:
- HTTP 200 or a page-load event.
- Successful syntax, type, lint, or build checks.
- A DOM selector or component existing.
- A click handler being registered.
- A screenshot showing a control.
Functional evidence must exercise the user-visible result and compare it with the acceptance condition. Persistence evidence must cross a refresh or restart boundary. Import/restore evidence must compare record counts and representative fields before and after recovery.
Use `passed`, `failed`, `blocked`, or `not-run`. Never convert `blocked` or `not-run` into `passed`. A required failed, blocked, or not-run control prevents `verification-passed`.
## Required coverage
Scale tests to confirmed behavior, covering as applicable:
- Primary end-to-end workflow with empty and populated data.
- Create, edit, delete, filtering, search, and navigation.
- Persistence across refresh or restart.
- Export, clear, import, restore, merge, overwrite, and corrupt input.
- Desktop and phone layouts, text wrapping, overflow, and touch targets.
- Keyboard operation, focus visibility, labels, contrast, zoom, and reduced motion.
- Loading, error, permission-denied, offline, and integration failure behavior.
- Authentication, authorization, tenant isolation, secret handling, and rate limits for L3-L4.
- Startup, port conflicts, shutdown, backup creation, and restore for a selected local-service profile.
## State evidence
The direct-build state sequence is:
1. `spec-approved`: required decisions and acceptance conditions are recorded; approval evidence satisfies the level gate.
2. `ui-approved`: a new or materially changed UI is approved, or prior approval is explicitly inherited because no material visual change occurred.
3. `implementation-complete`: every confirmed in-scope item is implemented and mapped to controls or non-UI behavior.
4. `verification-passed`: all required tests have actual results and acceptable evidence.
Run `scripts/workflow_guard.py` at every transition. Do not claim the workbench is complete before the fourth state passes.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |