Token导航 LogoToken导航TokenDH.com
研究检索只读github未标认证来源可访问clear审计提醒

blog-writer博客作家

Agent Skill

blog-writer 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。

总安装

4,845

周安装

196

GitHub Stars

15

下载量

1,521
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

3

许可证

MIT

最后核验

2026-05-01

来源状态

来源可访问

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:blog-writer(博客作家)
来源仓库:https://github.com/lightfastai/lightfast
仓库路径:skills/blog-writer
安装命令:
npx skills add https://github.com/lightfastai/lightfast --skill blog-writer
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。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/lightfastai/lightfast --skill blog-writer

简介

blog-writer 用于查找、检索和筛选相关信息。

  • 适合在 Codex、Claude、Cursor、Gemini CLI 中根据关键词快速定位候选结果。
  • 通过 npx skills add 命令从指定仓库安装,需确认权限和维护状态。
  • 使用前建议核验是否会触发联网、命令执行或文件读写操作。
  • 适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

SKILL.md

Blog Writer

Create clear, accurate blog posts that help developers understand Lightfast capabilities and industry trends.

Critical: Accuracy Over Marketing

Before writing anything:

  1. Verify every claim:

- If you cite a number, confirm the source - If you mention a feature, confirm it exists in production - When uncertain, ask for clarification

  1. Never oversell:

- Disclose limitations: "Currently supports X; Y coming in vZ" - Be honest about beta status and rollout timelines

  1. Match category voice:

- Engineering: Technical authority, data-driven - Company: Visionary, category-defining - Product: Problem-solver, benefit-oriented - Tutorial: Step-by-step, practical - Research: Data-driven, methodology-first

Writing Guidelines

  1. Concise & scannable: Match category word counts
  2. Lead with value: Start with what readers gain
  3. Be transparent: Mention beta status, limitations
  4. Active voice: "You can now..." not "Users are able to..."
  5. No emoji: Professional tone
  6. Include TL;DR: 20-300 chars for AI citation (in frontmatter tldr field)
  7. FAQ section: 3-5 questions matching search queries (in frontmatter faq[])
  8. Code examples: Required for Engineering posts

Workflow

  1. Detect category from input or ask if unclear
  2. Load category style from resources/categories/{category}.md
  3. Research topic using web-search-researcher agent
  4. Draft following category template and templates
  5. Add AEO elements per aeo-requirements
  6. Review with checklist

Quick Reference

Category Selection

CategoryUse WhenAudience
EngineeringTechnical deep-dives, architecture, researchDevelopers, engineers
CompanyFunding, partnerships, events, hiringExecutives, investors
ProductFeature launches, updates, tutorialsCustomers, prospects
TutorialStep-by-step how-tos, guidesDevelopers learning
ResearchData analysis, findings, benchmarksEngineering leaders

Do

  • Include code examples (Engineering/Tutorial)
  • Use "shift from/to" narratives (Company)
  • Lead with pain point (Product)
  • Link to 3-5 related docs

Don't

  • Marketing buzzwords without substance
  • Claims without verification
  • Long paragraphs (keep sections scannable)
  • Generic corporate speak

Output

Save drafts to: thoughts/blog/{YYYY-MM-DD-slug}.md

Required Frontmatter Fields

Every draft MUST include (maps to BlogPostSchema in apps/www/src/lib/content-schemas.ts):

  • title (core)
  • description (150-160 chars — this is the meta description)
  • keywords[] (min 3 — first entry is primary keyword)
  • ogTitle, ogDescription, ogImage (social/OG)
  • authors[] (structured array with name, url, twitterHandle)
  • publishedAt, updatedAt (ISO datetimes)
  • category (engineering | product | company | tutorial | research)
  • readingTimeMinutes (integer)
  • featured (boolean)
  • tldr (20-300 chars)
  • faq[] (min 1 entry)
  • _draft: true (traceability)

See resources/templates.md for complete frontmatter template.

Resources

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

04

需要参考平台分布和安装热度时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Claude Code

27.94%
按下载量换算425

Gemini CLI

24.7%
按下载量换算376

Antigravity

19.41%
按下载量换算295

OpenCode

12.69%
按下载量换算193

windsurf

8.92%
按下载量换算136

Cursor

3.54%
按下载量换算54

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

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。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。

来源信息

继续浏览同类 Skills