一键导入
plan
Use when asked to plan a coding task. Produces a written implementation plan (research plus steps) for a code change.
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
Use when asked to plan a coding task. Produces a written implementation plan (research plus steps) for a code change.
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
CLI tool for interacting with MCP (Model Context Protocol) servers. Use when invoking MCP tools, managing MCP servers, or working with MCP-based APIs. Triggers on "MCP server", "mcpli", "invoke MCP tool", or any MCP server interaction.
Use when executing a plan or implementing a coding task. Turns a plan or clear request into working code, tests, and verification.
Use for researching, checking facts, looking things online, comparing options, etc.
Use before performing a rebase or resolving rebase conflicts.
Use when opening a pull request, pushing changes for review, or when the user says "open PR", "create PR", "send for review", or "get this merged".
Reflect on the current conversation to find ways to improve agent skills and config. Use for "retro", "retrospective", "reflect on this conversation", "what could be improved", or "extract a skill from this".
| name | plan |
| description | Use when asked to plan a coding task. Produces a written implementation plan (research plus steps) for a code change. |
Research the codebase and produce an implementation plan. Make no source-code changes.
A plan must be self-contained. Assume the reader has only the plan, with no access to the conversation that produced it. It must carry enough context and rationale for a fresh agent to make good decisions on its own: key constraints, decisions already made and why, and relevant background found during research. It does not need exhaustive detail or exact code locations; it needs enough that a competent agent can find the rest and choose well. When deciding whether something belongs in the plan, apply this test: could a fresh agent build the right thing from the plan alone?
Before planning, explore the codebase to understand what exists:
Choose a detail level based on complexity:
Minimal, for simple, well-understood changes:
Comprehensive, for architectural changes, complex features or entire new projects:
Default to minimal.
For "Skills to use", recommend the skills available in your context that match the work in the plan. List them by name, each with a one-line note on when it applies during implementation (e.g. "tdd — for the parser changes", "git-commit — when committing"). The implementing agent loads them itself.
Present the plan directly in your response.