免费在 GitHub 公开

wechat-title-strategist

Evaluate, rewrite, and select WeChat Official Account article titles for open rate, depth, reader promise, and account tone. Use when drafting, reviewing, publishing, or revising公众号文章标题, especially when titles may be generic, stale, clickbait, AI-flavored, or misaligned with the article's commercial/editorial intent.

中文简介正在准备中。当前显示原文说明。

EthanYoQ/Skill-hub

实测评分未实测

查看安装方法

包含的文件（1）

SKILL.md2.7 KB

SKILL.md（原文）

安装前可以查看交给 AI 代理的指令内容。

WeChat Title Strategist

Use this skill before finalizing any WeChat article title.

Core Standard

A title must pass at least one primary route:

Depth route: expresses a non-obvious thesis, business insight, structural judgment, or contrarian read.
Open-rate route: creates a concrete curiosity gap, reader pain, useful promise, or timely tension that makes the target reader want to open.

If a title passes neither route, reject it even if it is accurate.

Hard Rejection Rules

Reject titles that:

Lead with stale AI discourse such as "别再写 Prompt", "Prompt 已死", "AI 时代来了", unless the article's actual evidence makes that wording newly valuable.
Use vague academic wording: "浅谈", "思考", "探索", "实践", "趋势分析", "未来已来".
Sound like generic AI slop: "一文看懂", "重磅", "颠覆", "震撼", "必须收藏", without concrete payoff.
Contain too many English terms for a general WeChat audience when Chinese wording would be clearer.
Describe the topic but not the reader's reason to open.
Promise a conclusion the article does not prove.

Title Test

For each candidate, score quickly:

Reader: who is supposed to care?
Promise: what will they get after opening?
Tension: what conflict, surprise, or stakes exist?
Evidence: what concrete fact, data point, product, or scene supports it?
Tone: does it raise or lower the account's perceived quality?

Reject if any answer is vague.

Generation Workflow

Extract the article's strongest thesis, reader pain, evidence, and account intent.
Generate 8-12 candidates across these buckets:
- Depth thesis.
- Business implication.
- Concrete evidence or number.
- Reader pain / workplace scene.
- Contrarian but accurate angle.
- Straight declarative title.
Remove weak or stale candidates using the rejection rules.
Return the top 3-5 with one-line rationale and risk.
Pick one final title only after it passes the title test.

Output Format

## Title Diagnosis
- Current title: ...
- Verdict: pass/fail
- Main problem: ...

## Candidates
1. ... — [route] [why it works] [risk]

## Recommended
...

Keep titles mostly under 30 Chinese characters when possible. A longer title is acceptable only when the extra words add real specificity.

评价

还没有评价。欢迎分享你的使用感受。

无需登录即可发表。欢迎分享使用感受、提问，或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

同一仓库的技能

aact-bulk-trials

免费

使用 AACT (Aggregate Analysis of ClinicalTrials.gov) PostgreSQL 数据仓库进行批量、历史、聚合性临床试验数据挖掘。Use this skill when the user requests bulk SQL analysis over the full clinical trials data warehouse — historical trial trends, disease landscapes, similar-design matching, or multi-year aggregations across hundreds of thousands of NCT records. 触发场景包括：AACT 查询、临床试验批量分析、PostgreSQL 试验数据、全量 NCT 检索、试验数据挖掘、历史试验分析、clinical trials data warehouse、SQL trials、bulk trial analysis、disease landscape、试验设计相似性匹配、跨年度聚合、sponsor/phase/country 多维统计。**与 clinical-trials-v2 差异**：本 skill 走批量 SQL · 离线大数据（PostgreSQL）；v2 走实时 API · 单查询。两者互补：单条 NCT 实时状态用 v2，百万级历史挖掘用本 skill。支持云端公共 PostgreSQL（aact-db.ctti-clinicaltrials.org · 零部署）和每日 dump 本地还原（高性能 · 离线）两种连接方式，自动检测优先用本地。跨平台（macOS/Linux/Windows）参数化 SQL 防注入，read-only 强制保护。

中文简介正在准备中。当前显示原文说明。