mcpField
K
krypton-planning

krypton-planning

github.com/jturntdevAgent 工作流13262 次安装17 周前更新
代码仓库

关于 krypton-planning

Krypton Planning 是将需求整理为目标计划包的 Agent 技能,面向编码前需要明确目标、事实归属、边界和验收方式的团队。它先梳理架构切片与切换路径,再拆分任务并要求计划审查。

主要功能

  1. 01定义目标、现状与预期结果
  2. 02明确事实归属与契约边界
  3. 03梳理文件范围和读写路径
  4. 04生成 PLAN.md、GOAL.md 与 EVIDENCE.md
  5. 05拆分任务并进行计划审查

适用场景

  1. 01把模糊需求转成可执行计划
  2. 02规划替换旧行为时的切换与退出条件
  3. 03为实现任务指定验证命令和验收证据

安装

复制下面的命令,在终端中运行。

$npx skills add https://github.com/jturntdev/krypton --skill krypton-planning

安装后在 Claude Code 中输入/krypton-planning确认已加载。

附带文件agents/openai.yamlplan-reviewer-prompt.md

详细介绍

Krypton Planning

Krypton Planning 会将请求转化为目标包,在编写代码之前明确预期结果、事实责任方、契约、切换和验收证据。

核心规则

不要把任务列表当作计划。只有回答了以下问题,Krypton 计划才算准备就绪:

What outcome are we serving?
What current behavior is being replaced, redirected, deleted, or demoted?
Who owns the truth?
What contract crosses the boundary?
What is the smallest high-value slice?
What proves the result from the target person's perspective?
What kill criteria stop duplicate paths from living forever?

如果责任归属、切换、契约或证据不明确,请停止并先梳理,再制定计划。

工作流

  1. 编写结果契约:
Plan title:
Intent:
Current behavior:
Expected outcome:
Target-perspective output:
Truth owner:
Contract boundary:
Cutover:
Displaced path:
Value density:
Evidence gate:
Acceptance evidence:
Evidence lane:
Kill criteria:
Non-goals:
Risk if wrong:
  1. 在列出任务之前梳理架构切片:
Files to create:
Files to modify:
Files to avoid:
Source of truth:
Read path:
Write path:
Contract boundary:
Integration points:
Migration/cutover:
Displaced path:
Acceptance evidence gate:

对于范围较大或不明确的代码仓库,如果运行环境支持 Agent,请安排一名只读探索者。提出一个范围明确的问题,例如“梳理事实来源、读写路径、不安全文件和证据关卡”。执行会话应使用这份梳理结果,而不是重新探索同一个切片。

  1. 除非用户指定其他位置,否则保存目标包:
docs/goals/<goal-slug>/PLAN.md
docs/goals/<goal-slug>/GOAL.md
docs/goals/<goal-slug>/EVIDENCE.md

PLAN.md 包含完整计划。GOAL.md 是简短的执行提示词,指向计划而不是复制计划。EVIDENCE.md 初始为空,并在执行期间填入从目标视角捕获的实际证明。

  1. 按以下格式开始编写 PLAN.md 的头部:
# [Outcome Title] Implementation Plan

**Intent:** ...
**Current Behavior:** ...
**Expected Outcome:** ...
**Target-Perspective Output:** ...
**Truth Owner:** ...
**Contract Boundary:** ...
**Cutover:** ...
**Displaced Path:** ...
**Value Density:** ...
**Evidence Gate:** ...
**Acceptance Evidence:** ...
**Evidence Lane:** ...
**Kill Criteria:** ...
**Architecture Slice:** ...
**Plan Review Gate:** Requires PRE review before execution.
  1. 将工作拆分为小任务。每项任务都要注明具体文件、允许的范围、预期输出、验证命令、验收证据,以及是否可以并行执行。

  2. 尽可能进行 PRE 计划审查。使用 plan-reviewer-prompt.md 作为单项提示词模板。修复阻塞性问题或明确征得用户同意之前,不要执行。

GOAL.md 格式

使用这份精简的交接内容。操作人员应将其作为 /goal 提示词或下一会话的提示词粘贴到 Codex 或 Claude 中:

# Goal: [Outcome Title]

Use Krypton Execution to execute `docs/goals/<goal-slug>/PLAN.md`.

Core rules:
- Treat PLAN.md as the source plan.
- Preserve intent, ownership, contract, cutover, evidence, and kill criteria.
- Do not add a new dominant path without deleting, redirecting, demoting, or shimming the displaced path.
- Capture acceptance evidence from the target perspective and record it in EVIDENCE.md.
- Say "implemented but unproven" if that evidence cannot be captured.

EVIDENCE.md 格式

创建目标包时同时创建此文件,并在执行期间更新:

# [Outcome Title] Evidence

## Acceptance Evidence

Record the real artifact that proves the outcome from the target perspective.

## Verification

Record the focused check that passed, including the command and relevant output.

## Review Notes

Record reviewer, maintainer, or verifier findings that changed the result.

评论

支持 Markdown