plan

仓库创建 2026年7月5日最近提交 22 天前SkillHot 收录 22 天前

▸ 精选理由

有助于明确边界与假设，减少返工并支持审查。

这个 Skill 做什么

探索代码库并输出可验证、独立编号的实现计划与风险假设。

先遍历 codebase，把实现拆成可验证、独立编号的任务，并把所有假设标注为 SAFE 或 RISKY；那些难回滚的 RISKY 决策会先变成问题请你确认。适合在开始编码或估算前把工作和风险说清楚。结果是既能当执行清单又能当风险合同，便于追踪和验收。

▸ 展开 SKILL.md 英文原文

Explore the codebase and produce an implementation plan with numbered, independently verifiable tasks. Surfaces every assumption (SAFE/RISKY); RISKY + hard-to-reverse decisions come back as questions before the plan is final.

开发编程计划代码探索任务拆解通用

Stars

Forks

仓库内 Skill

7 日增星

安装 / 使用

给你的 Agent 一句话（通用）

帮我安装这个 skill：https://raw.githubusercontent.com/eduardkumskyi/octo/master/skills/plan/SKILL.md

或 curl 直取 SKILL.md

curl -fsSL "https://raw.githubusercontent.com/eduardkumskyi/octo/master/skills/plan/SKILL.md"

SKILL.MD 节选查看完整文件 ↗

## Progress Contract

Register these steps as a native task list at Step 2, before beginning exploration.
Report progress as "N steps remaining, size class S/M/L" — never wall-clock ETAs.

Register steps in the native task list named `🐙 <n>/<total> — <step name>`; update each to in_progress/completed as you go — the checklist is the user's primary progress view.

Steps: (1) read-project-context, (2) register-progress, (3) explore,
(4) author-plan, (5) plan-self-review, (6) assumption-gate, (7) save-and-register.

## Arguments

- **`<task description>`** — what to plan. If `.claude/octo/specs/` contains a spec for this
  work (or the user names one explicitly), the plan **MUST** consume it:

via SKILL·HOT · 数据来自 GitHub 公开信息 · 原文版权归作者所有