Add comprehensive AI prompt system for protocol development

[Copilot]

Establishes structured AI context for GitHub Copilot and custom assistants across all ObjectStack protocol layers.

Architecture

Protocol Layer Prompts (5 files, ~2,590 lines)

• data-protocol.prompt.md - ObjectQL: Fields, Objects, Validation, Permissions, Workflows, Query AST
• ui-protocol.prompt.md - ObjectUI: Views, Apps, Dashboards, Reports, Themes, Widgets
• system-protocol.prompt.md - ObjectOS: Manifests, Plugins, Drivers, Identity, RBAC, i18n
• ai-protocol.prompt.md - AI integration: Agents, Tools, RAG, Conversations, Model configs
• api-protocol.prompt.md - API contracts: Envelopes, Requests, Error codes, REST/GraphQL

Development Role Prompts (3 files, ~1,457 lines)

• testing-engineer.prompt.md - Schema validation, edge cases, coverage patterns
• documentation-writer.prompt.md - TSDoc, concept guides, API references, tutorials
• example-creator.prompt.md - Full apps, feature demos, real-world scenarios

Navigation (3 files)

• README.md / README.zh-CN.md - Categorized index with quick commands
• OVERVIEW.md - Visual guide and statistics

Structure

Each prompt defines:

• Role context and scope
• Core responsibilities with Zod schema patterns
• Quick interaction commands (e.g., "Create Field Protocol" → complete implementation)
• Coding standards: camelCase config keys, snake_case machine names, Zod-first approach
• 100+ runnable code examples

Example Usage

// Prompt auto-loads when editing packages/spec/src/data/field.zod.ts
// AI context includes:
// - All 23+ field types (text, lookup, formula, etc.)
// - Type-specific properties (reference, cascade, expression)
// - Naming conventions and validation patterns
// - Integration with Object schema

export const FieldSchema = z.object({
  name: z.string().regex(/^[a-z_][a-z0-9_]*$/),  // snake_case enforced
  type: FieldTypeSchema,
  reference: z.string().optional(),  // Required for lookup type
  // ... complete pattern provided by prompt
});

Prompts integrate with GitHub Copilot's automatic context loading based on file path and task type.

Original prompt

帮我生成所有需要的AI提示词

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Review	Updated (UTC)
spec	Error		Jan 21, 2026 3:23pm

[github-actions]

This PR is very large. Consider breaking it into smaller PRs for easier review.