plan

GitHub

该技能用于在用户明确要求时,基于仓库现有架构和代码证据,创建结构化的实施计划。它强调通过垂直切片分解任务,明确决策依据、范围、契约及验证标准,确保计划具备可执行性且不与现有规划体系冲突。

.agents/skills/plan/SKILL.md stella/stella

触发场景

用户明确要求制定实施计划 需要基于证据拆解功能实现步骤

安装

npx skills add stella/stella --skill plan -g -y
更多选项

非标准路径

npx skills add https://github.com/stella/stella/tree/main/.agents/skills/plan -g -y

不安装直接使用

npx skills use stella/stella@plan

指定 Agent (Claude Code)

npx skills add stella/stella --skill plan -a claude-code -g -y

安装 repo 全部 skill

npx skills add stella/stella --all -g -y

预览 repo 内 skill

npx skills add stella/stella --list

SKILL.md

Frontmatter
{
    "name": "plan",
    "description": "Create a concise, evidence-backed implementation plan in the repository planning area when the user explicitly asks for a plan."
}

Plan

Create an implementation plan only when the user explicitly requests one. A plan records decisions and executable slices; it is not a substitute for inspecting the repository.

1. Establish the Planning Context

Read applicable repository instructions and the existing planning system. Prefer the repository's established location and format. When present, inspect:

  • .agents/ARCHITECTURE.md
  • .agents/GOALS.md
  • .agents/STATUS.md
  • recent related plans
  • the code, tests, schemas, and configuration the task would change

Use $ARGUMENTS as a short slug when provided; otherwise derive one from the task.

Do not create a second planning system. Use .agents/plans/ only when it is already established or the repository clearly adopts the shared convention. If no planning area exists and the repository does not adopt this convention, ask the user where to save the plan before creating a new directory.

2. Resolve Decisions With Evidence

Trace the current behavior through real entry points and boundaries. Record facts separately from proposals. Ask about a materially different product, security, migration, or compatibility choice only when repository evidence cannot resolve it. Do not pause for discoverable implementation details.

Prefer vertical slices that leave the repository working after each slice when the implementation shape is sufficiently settled. Otherwise plan outcomes and contracts without inventing files or symbols. Identify ownership boundaries, data contracts, invalid states, rollout risks, and generated artifacts explicitly.

3. Create a Collision-Safe File

Follow the repository's existing naming scheme when one exists and is safe for concurrent worktrees. Otherwise use:

YYYYMMDD-HHMMSS-<slug>-<short-unique-suffix>.md

Use UTC for the timestamp. Create the file exclusively and retry with a new suffix on collision. Never overwrite an existing plan. Do not derive a global sequential number from a directory listing: concurrent worktrees can choose the same number.

4. Write the Plan

Use only sections that carry information:

# Plan: [Feature]

## Goal

What outcome changes, for whom, and why.

## Current State

Relevant behavior, entry points, constraints, and evidence.

## Decisions

- **Decision**: choice, alternatives considered, and why they were rejected.

## Scope

In scope and intentionally out of scope.

## Vertical Slices

1. End-to-end slice with real files, boundaries, and a verifiable outcome.

## Contracts and Invariants

Types, schemas, authorization, compatibility, idempotency, and failure behavior.

## Verification

Focused checks, integration coverage, and observable acceptance criteria.

## Rollout and Recovery

Migration, sequencing, monitoring, rollback, or "Not applicable".

## Open Questions

Only unresolved decisions that can change the plan.

Name concrete files and symbols only where repository evidence and settled implementation decisions support them. Avoid pseudocode unless a contract would otherwise remain ambiguous. Keep the plan concise enough to stay useful during implementation.

Report the created path and any decision that still needs the user.

版本历史

  • dd81665 当前 2026-08-16 07:08

    从使用顺序编号改为时间戳加唯一后缀的防冲突命名;强调基于代码证据而非假设;新增验证、契约和回滚章节。

  • 85792bd 2026-07-24 16:11

同 Skill 集合

.agents/skills/click-around/SKILL.md
.agents/skills/conventions-ai/SKILL.md
.agents/skills/conventions-db/SKILL.md
.agents/skills/conventions-i18n/SKILL.md
.agents/skills/conventions-ingestion/SKILL.md
.agents/skills/conventions-perf/SKILL.md
.agents/skills/conventions-scale/SKILL.md
.agents/skills/conventions-security/SKILL.md
.agents/skills/conventions-use-effect/SKILL.md
.agents/skills/conventions-ux/SKILL.md
.agents/skills/dev/SKILL.md
.agents/skills/finish-pr/SKILL.md
.agents/skills/new-handler/SKILL.md
.agents/skills/open-pr/SKILL.md
.agents/skills/product-deep-think/SKILL.md
.agents/skills/product-think/SKILL.md
.agents/skills/rabbit-round/SKILL.md
.agents/skills/regression-hunt/SKILL.md
.agents/skills/security-audit/SKILL.md
.agents/skills/update-deps/SKILL.md
.ai/local-skills/click-around/SKILL.md
.ai/local-skills/conventions-ai/SKILL.md
.ai/local-skills/conventions-db/SKILL.md
.ai/local-skills/conventions-i18n/SKILL.md
.ai/local-skills/conventions-ingestion/SKILL.md
.ai/local-skills/conventions-perf/SKILL.md
.ai/local-skills/conventions-scale/SKILL.md
.ai/local-skills/conventions-security/SKILL.md
.ai/local-skills/conventions-use-effect/SKILL.md
.ai/local-skills/conventions-ux/SKILL.md
.ai/local-skills/dev/SKILL.md
.ai/local-skills/new-handler/SKILL.md
.ai/local-skills/open-pr/SKILL.md
.ai/local-skills/plan/SKILL.md
.ai/local-skills/product-deep-think/SKILL.md
.ai/local-skills/rabbit-round/SKILL.md
.ai/local-skills/security-audit/SKILL.md
.ai/local-skills/update-deps/SKILL.md
packages/cli/skills/stella-cli/SKILL.md
packages/skills/blueprints/answer-from-sources/SKILL.md
packages/skills/blueprints/blank/SKILL.md
packages/skills/blueprints/check-against-rules/SKILL.md
packages/skills/blueprints/intake-to-draft/SKILL.md
.agents/skills/conventions-testing/SKILL.md
.ai/local-skills/conventions-testing/SKILL.md

元信息

文件数
0
版本
f4b61c7
Hash
ecd60dee
收录时间
2026-07-24 16:11

首页 - Wiki
Copyright © 2011-2026 iteam. Current version is 2.155.2. UTC+08:00, 2026-08-22 00:21
浙ICP备14020137号-1 $访客地图$