Loads stored development state to resume work from previous sessions.

Install

mkdir -p .claude/skills/load && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/14197" && unzip -o skill.zip -d .claude/skills/load && rm skill.zip

Installs to .claude/skills/load

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

恢复开发进度。读取 progress.json、context.md、tasks.md,输出恢复报告并等待确认。
56 charsno explicit “when” trigger
Beginner

Key capabilities

  • Read progress.json to get current task details
  • Read context.md for development context snapshots
  • Read tasks.md for task lists and completion status
  • Read plan.md for design decision background
  • Output a recovery report summarizing progress

How it works

The skill reads specific project tracking files (progress.json, context.md, tasks.md, and optionally plan.md) to gather information about the current development status. It then compiles this information into a recovery report for the user.

Inputs & outputs

You give it
Execution of the /load command
You get back
A recovery report and a prompt for user confirmation

When to use load

  • Resuming a coding session
  • Reviewing previous development progress
  • Syncing state after a system interruption
  • Verifying current task completion status

About this skill

/load 命令 — 恢复开发进度

用法

  • /load — 读取存档,恢复到上次进度点

参数:无


执行流程

[BLOCKING CONSTRAINT] 以下每个步骤都必须使用 Read 工具实际读取文件,禁止凭记忆回答,禁止跳过任何文件。

步骤 1:读取 progress.json

使用 Read 工具读取 specs/progress.json,获取:

  • feature — 当前需求名称
  • branch — 开发分支
  • worktree — worktree 路径(如有)
  • specsDir — specs 目录路径
  • stage / stageLabel — 当前阶段
  • currentTask — 当前任务
  • completedTasks / totalTasks — 任务进度
  • updatedAt — 上次更新时间

如果文件不存在,输出错误并终止:

[/load] 错误:未找到 specs/progress.json,当前没有进行中的开发任务

步骤 2:读取 context.md

[MUST] 使用 Read 工具读取 specs/<feature>/context.md

这是最关键的文件,包含上次保存的开发上下文快照。重点关注最新一条快照中的:

  • "未完成(下次继续)" — 这是恢复后第一件要做的事
  • "当前状态" — 了解上次做到了哪一步
  • "关键决策" — 确保延续之前的技术决策
  • "遗留问题" — 检查是否有需要先处理的问题

如果文件不存在,记录警告但不终止(可能是首次开发)。

步骤 3:读取 tasks.md

[MUST] 使用 Read 工具读取 specs/<feature>/tasks.md

获取完整的任务列表和完成状态,确认:

  • 哪些任务已完成([x]
  • 当前任务的具体步骤和验收标准
  • 下一个待执行任务

步骤 4:读取 plan.md(可选)

使用 Read 工具读取 specs/<feature>/plan.md

获取设计决策背景,帮助理解整体架构。如果文件不存在,跳过此步骤。

步骤 5:输出恢复报告

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
  进度已恢复
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

需求:<feature>
分支:<branch>
阶段:<stageLabel>
任务进度:<completedTasks>/<totalTasks>

上次停在:
  - 任务:Task <id> - <title>
  - 进度:<从 context.md 最新快照提取>
  - 时间:<updatedAt>

关键上下文:
  - <context.md 中最新快照的关键决策摘要>

下一步:
  - <从 context.md "未完成(下次继续)"提取的待办事项>

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

步骤 6:等待用户确认

输出恢复报告后,必须等待用户确认再继续开发。询问:

已恢复到上次进度点。确认后继续开发,或告诉我需要调整的地方。

自检清单

[BLOCKING] 输出恢复报告前,逐项确认:

  • 是否使用 Read 工具实际读取了 progress.json?(不是凭记忆)
  • 是否使用 Read 工具实际读取了 context.md?
  • 是否使用 Read 工具实际读取了 tasks.md?
  • 恢复报告中的"下一步"是否来自 context.md 而非推测?

如有未完成项,立即补做后再输出报告。

When not to use it

  • When no development task is in progress
  • When files like progress.json, context.md, or tasks.md are not present

Limitations

  • Requires the existence of specs/progress.json to proceed
  • Requires the existence of specs/<feature>/context.md and specs/<feature>/tasks.md

How it compares

This workflow provides a structured way to resume development by explicitly reading and summarizing project state from predefined files, unlike manually recalling or searching for progress information.

Compared to similar skills

load side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
load (this skill)01moNo flagsBeginner
pptx3936moReviewAdvanced
nano-pdf633moReviewBeginner
video-downloader1017moReviewBeginner

Try saying

Example prompts that trigger this skill in your AI assistant.

You might also like

pptx

anthropics

Presentation creation, editing, and analysis. When Claude needs to work with presentations (.pptx files) for: (1) Creating new presentations, (2) Modifying or editing content, (3) Working with layouts, (4) Adding comments or speaker notes, or any other presentation tasks

393763

nano-pdf

openclaw

Edit PDFs with natural-language instructions using the nano-pdf CLI.

63300

video-downloader

ComposioHQ

Downloads videos from YouTube and other platforms for offline viewing, editing, or archival. Handles various formats and quality options.

101255

youtube-transcript

michalparkola

Download YouTube video transcripts when user provides a YouTube URL or asks to download/get/fetch a transcript from YouTube. Also use when user wants to transcribe or get captions/subtitles from a YouTube video.

68277

using-superpowers

obra

Use when starting any conversation - establishes mandatory workflows for finding and using skills, including using Skill tool before announcing usage, following brainstorming before coding, and creating TodoWrite todos for checklists

95205

browser-automation

browserbase

Automate web browser interactions using natural language via CLI commands. Use when the user asks to browse websites, navigate web pages, extract data from websites, take screenshots, fill forms, click buttons, or interact with web applications. Triggers include "browse", "navigate to", "go to website", "extract data from webpage", "screenshot", "web scraping", "fill out form", "click on", "search for on the web". When taking actions be as specific as possible.

39230

Search skills

Search the agent skills registry