360 Secure SkillHub
360 安全技能中心

Linkedin Post

这是一款命令行LinkedIn内容全流程管理工具,核心功能涵盖帖子起草、编辑、优化、调度,还能生成钩子、话题标签、CTA,支持改写、翻译、语气调整等操作。典型场景包括快速记录帖子草稿、打磨吸睛开头、规划发布日程、迭代内容质量。所有操作生成带时间戳的日志,支持查看历史、统计数据,可导出为JSON/CSV/TXT格式,无需外部依赖。

ckchzh
ckchzh
热度 71.17
最近更新 2026-05-11 17:30:24
分类:营销依赖:纯提示词应用场景:文案创作 / 社交媒体 / 日程安排 / 任务管理

安装技能

复制下面的话术发给你的 Agent,它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill: npx secure-skills-cli add linkedin-post
Metadata
version
2.0.0
name
linkedin-post
description
LinkedIn文案生成、开头Hook、热门话题标签、轮播内容规划、高质量评论、个人简介优化。LinkedIn post writer with hooks, hashtags, carousel planning, comment templates.
author
BytesAgain
homepage
https://bytesagain.com
source
https://github.com/bytesagain/ai-skills

LinkedIn Post

Content toolkit for drafting, editing, optimizing, scheduling, and managing LinkedIn posts from the command line. Generate hashtags, craft hooks, write CTAs, rewrite content, translate text, adjust tone, create headlines, and build outlines — all with timestamped logging and full export capabilities.

Commands

Command Description
linkedin-post draft <input> Draft a new post (view recent drafts with no args)
linkedin-post edit <input> Log an edit entry (view recent edits with no args)
linkedin-post optimize <input> Log an optimization note (view recent with no args)
linkedin-post schedule <input> Schedule a post (view recent schedules with no args)
linkedin-post hashtags <input> Log hashtag ideas (view recent with no args)
linkedin-post hooks <input> Log opening hooks (view recent with no args)
linkedin-post cta <input> Log call-to-action ideas (view recent with no args)
linkedin-post rewrite <input> Log a rewrite (view recent rewrites with no args)
linkedin-post translate <input> Log a translation (view recent with no args)
linkedin-post tone <input> Log tone adjustments (view recent with no args)
linkedin-post headline <input> Log headline ideas (view recent with no args)
linkedin-post outline <input> Log post outlines (view recent with no args)
linkedin-post stats Show summary statistics across all log files
linkedin-post export <fmt> Export all data (json, csv, or txt)
linkedin-post search <term> Search all logs for a keyword
linkedin-post recent Show the 20 most recent history entries
linkedin-post status Health check with version, entries, disk usage
linkedin-post help Show help message
linkedin-post version Show version (v2.0.0)

How Data Commands Work

Each content command (draft, edit, optimize, schedule, hashtags, hooks, cta, rewrite, translate, tone, headline, outline) operates in two modes:

  • With arguments — saves a timestamped entry (YYYY-MM-DD HH:MM|value) to the command's .log file and records the action in history.log
  • Without arguments — displays the 20 most recent entries from that command's log file

Utility Commands

  • stats — iterates all .log files in the data directory, counts entries per category, and shows total count, disk usage, and earliest activity timestamp
  • export <fmt> — exports all log data into json, csv, or txt format, saved to ~/.local/share/linkedin-post/export.<fmt>. Reports output file path and byte count.
  • search <term> — performs case-insensitive search across all log files, grouped by category
  • recent — shows the last 20 lines from history.log
  • status — health check displaying version, data directory, total entry count, disk usage, last activity, and OK status

Data Storage

All data is stored in ~/.local/share/linkedin-post/:

  • Each command has its own log file (e.g., draft.log, edit.log, hashtags.log, hooks.log, etc.)
  • history.log — centralized activity log recording every command invocation
  • Export files saved as export.json, export.csv, or export.txt
  • Entries use pipe-delimited format: YYYY-MM-DD HH:MM|value

Requirements

  • Bash (with set -euo pipefail)
  • Standard Unix utilities: date, wc, du, head, tail, grep, basename
  • No external dependencies or API keys required

When to Use

  1. Drafting LinkedIn posts — use linkedin-post draft to capture post ideas and first drafts as they come to you
  2. Crafting engaging hooks — log attention-grabbing opening lines with linkedin-post hooks and iterate until you find the best one
  3. Managing hashtag strategy — track and organize hashtag sets with linkedin-post hashtags for consistent post tagging
  4. Scheduling content calendars — use linkedin-post schedule to plan when posts go live and maintain a publishing cadence
  5. Iterating on post quality — use linkedin-post rewrite, linkedin-post tone, and linkedin-post optimize to refine content through multiple revisions

Examples

# Draft a new post
linkedin-post draft "5 lessons I learned from shipping my first SaaS product in 30 days"

# Log hook ideas for the post
linkedin-post hooks "I almost quit on day 14. Here's why I didn't."
linkedin-post hooks "Most founders get this wrong about MVPs..."

# Add hashtag ideas
linkedin-post hashtags "#SaaS #StartupLife #ProductLaunch #IndieHacker #BuildInPublic"

# Write a CTA
linkedin-post cta "Drop a 🔥 if you've shipped something this month. I'll check out your project."

# Schedule the post
linkedin-post schedule "Publish Tuesday 9am EST — peak LinkedIn engagement window"

# Rewrite with different tone
linkedin-post rewrite "Shorter version: cut intro, lead with the lesson, end with question"
linkedin-post tone "Switch from formal to conversational — more 'you' and 'I', fewer buzzwords"

# Create a headline
linkedin-post headline "From Zero to $10K MRR: What Actually Worked"

# Build an outline
linkedin-post outline "1) Hook: surprising stat 2) Problem 3) 3 key lessons 4) CTA question"

# Translate for different audience
linkedin-post translate "Spanish version for LATAM LinkedIn audience"

# View summary statistics
linkedin-post stats

# Export everything as JSON
linkedin-post export json

# Search for all posts about SaaS
linkedin-post search "SaaS"

# Check system status
linkedin-post status

# View recent activity
linkedin-post recent

LinkedIn Best Practices

  • First 3 lines are everything — the "see more" hook determines whether people engage
  • Dwell time matters — longer posts that keep people reading get boosted by the algorithm
  • Comments > Reactions — posts with comments rank higher than those with just likes
  • Avoid external links in post body — put URLs in the first comment instead
  • Carousel posts get 1.5–2× more reach than text-only posts
  • Use linkedin-post hooks to iterate on your opening lines before publishing

Tips

  • Run linkedin-post help to see all available commands
  • Call any command without arguments to review recent entries for that category
  • Use linkedin-post search <term> to find entries across all log files
  • Export data regularly with linkedin-post export json for backups
  • Combine draft → hooks → hashtags → cta → optimize as a content creation pipeline
  • All data lives in ~/.local/share/linkedin-post/ — easy to back up or sync

Powered by BytesAgain | bytesagain.com | hello@bytesagain.com