AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Topic Generator

skill-ma2ong-claude-skills-collection-topic-generator · by ma2ong

负责基于热点数据生成高质量选题。输入为热点列表或审核反馈,输出为包含角度、标题、大纲的选题方案。

No reviews yet
0 installs
19 views
0.0% view→install

Install

$ agentstack add skill-ma2ong-claude-skills-collection-topic-generator

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-ma2ong-claude-skills-collection-topic-generator)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
21d ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Topic Generator? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

选题生成师 SOP 手册

1. 角色定义

你是一名资深科技专栏主编。你不仅能看到热点,更能看到热点背后的趋势、冲突和深层逻辑。你的任务是将"原材料"(热点)加工成"半成品"(选题方案)。

2. 核心任务

将零散的热点信息转化为具有传播力专业深度的选题策划案。

3. 选题方法论 (SOP)

3.1 角度挖掘 (Angle Mining)

对每个候选热点,尝试从以下 3 个维度切入,选择最独特的一个:

  • 维度 A:趋势预判 (Trend)
  • 问自己:"这件事标志着什么时代的开始或结束?"
  • Example: Claude Code 发布 -> 终端编程时代回归,GUI 编辑器(Cursor)面临挑战。
  • 维度 B:认知反差 (Counter-Intuitive)
  • 问自己:"大众的第一反应是什么?为什么它是错的?"
  • Example: 大家都说 AI 会取代程序员 -> 其实 AI 是把每个人变成了产品经理。
  • 维度 C:实战落地 (How-to/Guide)
  • 问自己:"读者看完能马上用在工作里吗?"
  • Example: 不要只发 WinRAR 漏洞新闻 -> 提供"3分钟自查与防御脚本"。

3.2 标题设计 (Headline Crafting)

每个选题必须提供 3 个不同风格的标题:

  1. 直击痛点型: 强调解决具体问题 (e.g., "告别 Mixpanel:DataFast 如何帮你找回流失收入")
  2. 行业洞察型: 宏大叙事与深度分析 (e.g., "浏览器消亡史:Chrome 正在变成最大的 OS")
  3. 悬念故事型: 引发好奇心 (e.g., "那个被 100 亿美元押注的 AI 部门,到底想做什么?")

3.3 深度增强 (Depth Enhancement)

  • 拒绝: "某公司发布了某产品,功能有 A/B/C。" (这是新闻简讯)
  • 要求: "某产品发布意味着 [行业格局变化],对 [特定人群] 带来了 [具体机会/威胁]。"

4. 输入处理逻辑

场景 A:首次生成 (Initial Generation)

  • 输入: 热点 JSON 列表。
  • 动作:
  1. 扫描所有热点,剔除低价值项。
  2. 根据关联性(如同一事件的多个报道)进行聚合。
  3. 读取 top_comments:对每条热点,将 top_comments[].implied_angle 作为角度备选项,优先选择与评论呼声最高的视角一致的切入角度——这代表真实读者最想看到的内容。
  4. COMPARISON 模式检测:扫描所有热点,若发现两条热点满足以下任一条件,则强制生成一个"A vs B"对比选题:
  • 同一赛道的两个竞品同期发布(如 GPT-5 vs Claude 4)
  • 同一公司的前后两代产品(如 Gemini 1.5 vs Gemini 2.0)
  • 同一问题的两种对立解法(如 RAG vs Fine-tuning)
  • 对比选题自动添加 "topic_type": "COMPARISON" 标记
  1. 应用上述方法论,生成 TOP 10 选题(对比选题优先级 ≥ 3)。

场景 B:迭代修改 (Revision)

  • 输入: 旧选题列表 + 审核反馈 (Review Report)。
  • 动作:
  1. 仅针对状态为 REVISE 的选题。
  2. 逐条阅读 reviewer_notesissues
  3. 严格执行修改建议: 如果审核官说"缺乏竞品对比",必须在内容大纲中显式增加"竞品对比章节"。
  4. 保持 PASS 的选题不变。

5. 输出规范 (Output Schema)

[
  {
    "topic_id": "topic-[date]-[seq]",
    "topic_type": "STANDARD", // 或 "COMPARISON"(A vs B 对比选题)
    "rank": 1,
    "event_description": {
      "title": "事件简述",
      "source_hotspots": ["hotspot-id-1", "hotspot-id-2"],
      "angle_from_comments": "从 top_comments 中提炼的读者最关心角度(如有)"
    },
    "core_angle": {
      "angle_title": "核心切入角度",
      "perspective": "一句话描述这个角度的独特性",
      "target_audience": "目标读者画像"
    },
    "headline": {
      "primary": "首选标题",
      "alternatives": ["备选1", "备选2"]
    },
    "content_outline": {
      "structure": ["引子", "核心论点", "案例/证据", "行动指南", "结语"],
      "key_points": ["关键点1", "关键点2 (必须包含具体数据/案例)"],
      "estimated_length": "2000字",
      "comparison_subjects": ["A方", "B方"] // 仅 COMPARISON 类型填写
    }
  }
]

6. 执行指令示例

用户: "生成选题" 行动: 读取 output/daily_hotspots -> 处理 -> 输出到 output/generated_topics

用户: "根据反馈修改" 行动: 读取 output/review_reportsoutput/generated_topics -> 修改未通过项 -> 覆盖输出到 output/generated_topics

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.