Token导航 LogoToken导航

Agent Skill · 研究检索

spec-kit-tasks(规格套件任务)

spec-kit-tasks 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

spec-kit-tasks 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为需要联网;许可证信息还不明确,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 69,087。

总安装 220周安装 9GitHub Stars 2下载 71核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:spec-kit-tasks(规格套件任务)
来源仓库:https://github.com/ahgraber/skills
仓库路径:skills/spec-kit-tasks
安装命令:
npx skills add https://github.com/ahgraber/skills --skill spec-kit-tasks
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/ahgraber/skills --skill spec-kit-tasks

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

spec-kit-tasks 用于查找、检索和筛选相关信息。适合在 Codex、Claude、Cursor、Gemini CLI 中根据关键词或任务线索快速定位候选结果。通过 npx skills add 命令从指定 GitHub 仓库安装,需结合原始 README 了解具体用法。

安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写操作。当前归类为研究检索类技能,适用于信息探索与筛选场景。

来源仓库内路径:skills/spec-kit-tasks

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/ahgraber/skills --skill spec-kit-tasks;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要需要联网,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

spec-kit-tasks 怎么安装?

在终端执行 npx skills add https://github.com/ahgraber/skills --skill spec-kit-tasks。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

spec-kit-tasks 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

spec-kit-tasks 需要什么权限?

权限级别是需要联网。该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

spec-kit-tasks 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

spec-kit-tasks 用的人多吗?

公开数据是累计安装 220 次,近一周安装 9 次,下载 71 次,GitHub 2 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

spec-kit-tasks 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
spec-kit-tasks

规格套件任务 · 本页

2202需要联网未明确通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Spec Kit Tasks

Generate an implementation-ready tasks.md from Spec Kit design artifacts.

Invocation Notice

  • Inform the user when this skill is being invoked by name: spec-kit-tasks.

When to Use

  • plan.md and spec.md exist and you need executable tasks for implementation.
  • tasks.md is missing, stale, or not aligned with current plan/spec artifacts (including post-spec-kit-reconcile updates).
  • You need user-story-scoped task phases with clear dependency and parallelization signals.

When Not to Use

  • The feature spec is missing (spec-kit-specify first).
  • High-impact ambiguity still blocks design decisions (spec-kit-clarify first).
  • Technical design artifacts are not finalized (spec-kit-plan first).
  • You are executing tasks rather than generating them (spec-kit-implement).

Router Fit

  • Primary route from spec-kit after spec-kit-plan.
  • Must complete before spec-kit-implement.
  • Supports spec-kit-analyze and spec-kit-reconcile by producing structured implementation intent.

Preconditions

  • Run from repository root (or a subdirectory inside it).
  • Active feature context resolves to one specs/<feature>/ directory.
  • plan.md exists and reflects the latest approved design.

Workflow

  1. Resolve feature paths and prerequisite gate:

- Run scripts/check-prerequisites.sh --json exactly once. - Parse FEATURE_DIR and AVAILABLE_DOCS. - Derive: - FEATURE_SPEC = FEATURE_DIR/spec.md - IMPL_PLAN = FEATURE_DIR/plan.md - TASKS = FEATURE_DIR/tasks.md - If spec.md is missing, stop and route to spec-kit-specify then spec-kit-plan.

  1. Load generation inputs:

- Required: plan.md, spec.md. - Optional (when present): research.md, data-model.md, contracts/, quickstart.md. - Template preference: - {REPO_ROOT}/templates/tasks-template.md - {REPO_ROOT}/.specify/templates/tasks-template.md - fallback: assets/tasks-template.md

  1. Extract planning context:

- From plan.md: stack, architecture, constraints, project structure. - From spec.md: prioritized user stories, acceptance criteria, independent test intent. - From optional docs: shared entities, external interfaces, and setup decisions.

  1. Build task phases:

- Phase 1: Setup. - Phase 2: Foundational prerequisites blocking all stories. - Phase 3+: one phase per user story in priority order. - Final phase: Polish/cross-cutting concerns.

  1. Generate tasks in strict checklist format:

- Required pattern: - [] T### [P?] [US#?] Action with file path. - Include [US#] only for user-story phases. - Include [P] only when tasks can run safely in parallel. - Include exact file paths in every implementation/test task description. - Add test tasks only when explicitly requested by spec/user.

  1. Validate coverage and ordering before writing:

- Every user story has independently testable tasks. - Dependencies are explicit and respect phase order. - Every task matches format requirements (checkbox, ID, labels, file path). - No orphaned entities/contracts without mapped tasks.

  1. Write tasks.md:

- Preserve heading order from the selected template. - Replace template placeholders and sample content with concrete feature tasks.

  1. Report completion:

- Absolute tasks.md path. - Total task count and per-story counts. - Parallel opportunities. - Suggested MVP slice (typically first priority story). - Readiness handoff for spec-kit-implement.

Task Format Guidance

Format Components

  1. Checkbox: ALWAYS start with - [] (markdown checkbox).
  2. Task ID: Sequential number (T001, T002, T003,...) in execution order.
  3. [P] marker: Include ONLY if the task is parallelizable (different files, no dependencies on incomplete tasks).
  4. [Story] label: REQUIRED for user story phase tasks only.

- Format: [US1], [US2], [US3], etc. (maps to user stories from spec.md) - Setup phase: no story label - Foundational phase: no story label - User story phases: MUST have story label - Polish phase: no story label

  1. Description: Clear action with exact file path.

Examples

  • CORRECT: - [] T001 Create project structure per implementation plan
  • CORRECT: - [] T005 [P] Implement authentication middleware in src/middleware/auth.py
  • CORRECT: - [] T012 [P] [US1] Create User model in src/models/user.py
  • CORRECT: - [] T014 [US1] Implement UserService in src/services/user_service.py
  • WRONG: - [] Create User model (missing Task ID and Story label)
  • WRONG: T001 [US1] Create model (missing checkbox)
  • WRONG: - [] [US1] Create User model (missing Task ID)
  • WRONG: - [] T001 [US1] Create model (missing file path)

Output

  • tasks.md under the active feature directory with dependency-ordered, story-scoped tasks.
  • Completion summary with task counts, dependency highlights, and MVP recommendation.

Key Rules

  • Organize tasks by user story so each story can be implemented and validated independently.
  • Keep task granularity implementation-ready: no vague one-liners and no giant umbrella tasks.
  • Preserve execution truth in IDs and dependency order (T001, T002,...).
  • Treat foundational work as blocking unless it is explicitly independent and parallel-safe.
  • Never invent tests by default; include test work only when requested.

Common Mistakes

  • Generating tasks from spec.md alone without using plan.md constraints.
  • Copying sample tasks from the template instead of replacing them.
  • Missing file paths or story labels in user-story phases.
  • Marking conflicting tasks as [P] even though they touch the same files/dependencies.
  • Producing tasks that cannot satisfy each story's independent test criteria.

References

  • references/spec-kit-workflow.dot for where task generation fits in the full Spec Kit sequence.
  • scripts/check-prerequisites.sh
  • assets/tasks-template.md
  • https://github.com/github/spec-kit/blob/9111699cd27879e3e6301651a03e502ecb6dd65d/templates/commands/tasks.md

平台分布

Codex

37.96%
按下载量换算27

Claude

30.38%
按下载量换算22

Cursor

17.88%
按下载量换算13

Gemini CLI

9.58%
按下载量换算7

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills