一键导入
project-planning
Create clear, step-by-step implementation plans with acceptance criteria using MCP helpers. EXCLUSIVE to planner agent.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Create clear, step-by-step implementation plans with acceptance criteria using MCP helpers. EXCLUSIVE to planner agent.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Brainstorm and advise on technical decisions using structured process and MCP helpers. EXCLUSIVE to brainstormer agent. Does NOT implement — only advises.
Diagnose errors and failing tests in Laravel + React + Python applications. Use when encountering bugs, exceptions, stack traces, 500 errors, TypeErrors, failing tests, or unexpected behavior. EXCLUSIVE to debugger agent.
Review code changes for correctness, security, performance, and maintainability. Use for PR reviews, code audits, pre-merge checks, or quality validation of Laravel + React + Python code. EXCLUSIVE to reviewer agent.
Plan and implement safe database schema changes including migrations, indexes, and backfills. Use when creating tables, adding columns, optimizing queries, or managing Eloquent/SQLAlchemy relationships. EXCLUSIVE to database-admin agent.
Manage deployment, Docker, CI/CD, server hardening, and infrastructure security. EXCLUSIVE to devops-engineer agent.
Keep project documentation accurate when behavior changes. EXCLUSIVE to project-manager agent.
| name | project-planning |
| description | Create clear, step-by-step implementation plans with acceptance criteria using MCP helpers. EXCLUSIVE to planner agent. |
| allowed-tools | Read, Grep, Glob, Bash, mcp_codex-bridge, mcp_gemini-bridge, mcp_context7, mcp_playwright, mcp_zread, mcp_web-search-prime, mcp_web-reader, mcp_zai-mcp-server, mcp_open-bridge |
Exclusive to: planner agent
mcp_gemini-bridge_consult_gemini(
query="Design architecture for [feature]: phases, risks, dependencies...",
directory="."
)
mcp_open-bridge_consult_gemini(
query="Design architecture for [feature]: phases, risks, dependencies...",
directory="."
)
mcp_context7_resolve-library-id(libraryName="laravel", query="[feature]")
mcp_context7_query-docs(libraryId="/laravel/docs", query="[specific pattern]")
PHP/Laravel: Laravel 12, Inertia.js, React 19, TypeScript, Tailwind Python: FastAPI, LangChain, LangGraph, Pydantic, pytest
One sentence summary of what we're building.
Create 3-5 phases, each independently testable.
| Size | Hours | Complexity |
|---|---|---|
| XS | 1-2 | Single file |
| S | 2-4 | Few files |
| M | 4-8 | Multiple files |
| L | 1-2 days | Multiple components |
| XL | 3-5 days | Full feature |
| Level | Criteria | Mitigation |
|---|---|---|
| 🔴 High | Data loss, security, breaking | Rollback plan, staging |
| 🟡 Medium | Performance, UX regression | Feature flag |
| 🟢 Low | Cosmetic, refactor | Standard testing |
# Phase N: [Name]
## Objective
[What this accomplishes]
## Tasks
- [ ] Task with file path
## Files
| File | Action |
|------|--------|
| `path/file` | Create/Modify |
## Verification
```bash
[commands]
[X hours]
## plan.md Output (REQUIRED)
Always create a consolidated `plan.md` file that contains ALL phases and context:
```markdown
# Plan: [Feature Name]
## Context
[What we're building and why]
## Code Patterns to Follow
[Key patterns from codebase research]
## Phases
| # | Name | Objective | Est. |
|---|------|-----------|------|
---
## Phase 1: [Name]
### Objective
[Goal]
### Tasks
- [ ] Task with `path/to/file`
### Files
| File | Action |
|------|--------|
---
[... repeat for each phase ...]
---
## Summary
- **Total Phases**: N
- **Estimated Effort**: X hours
- **Key Risks**: [list]