智能配文
caption
by BytesAgain
Generate captions for images, videos, and social media posts. Use when writing post captions, creating accessibility text, or batch-processing media.
安装
claude skill add --url github.com/openclaw/skills/tree/main/skills/ckchzh/caption文档
Caption
Caption is a content toolkit for drafting, editing, optimizing, and managing written content directly from the terminal. It tracks every entry with timestamps and supports searching, exporting, and reviewing your content history.
Commands
| Command | Description |
|---|---|
caption draft <text> | Save a draft entry (no args = show recent drafts) |
caption edit <text> | Save an edit entry (no args = show recent edits) |
caption optimize <text> | Save an optimization note (no args = show recent) |
caption schedule <text> | Schedule content for publishing (no args = show recent) |
caption hashtags <text> | Save hashtag sets (no args = show recent) |
caption hooks <text> | Save hook ideas (no args = show recent) |
caption cta <text> | Save call-to-action text (no args = show recent) |
caption rewrite <text> | Save rewritten content (no args = show recent) |
caption translate <text> | Save translated content (no args = show recent) |
caption tone <text> | Save tone-adjusted content (no args = show recent) |
caption headline <text> | Save headline variations (no args = show recent) |
caption outline <text> | Save content outlines (no args = show recent) |
caption stats | Show summary statistics across all categories |
caption export <fmt> | Export all data (formats: json, csv, txt) |
caption search <term> | Search across all logs for a term |
caption recent | Show the 20 most recent activities |
caption status | Health check — version, disk usage, entry counts |
caption help | Show all available commands |
caption version | Show version |
How It Works
Caption organizes content into separate log files by category (e.g., draft.log, hooks.log, tone.log). Each entry is timestamped and appended to the appropriate log. Running a command without arguments shows the 20 most recent entries for that category.
All actions are also recorded in history.log for a unified activity timeline.
Data Storage
All data is stored locally in ~/.local/share/caption/ by default:
draft.log,edit.log,optimize.log, etc. — Category-specific entry logshistory.log— Unified timestamped activity logexport.json,export.csv,export.txt— Generated export files
Each log entry uses the format: YYYY-MM-DD HH:MM|<content>
Requirements
- bash 4+ (uses
set -euo pipefailfor strict mode) - Standard Unix tools (
grep,wc,du,tail,head) - No API keys needed
- No external dependencies
When to Use
- Drafting social media captions — Use
caption draftto quickly log caption ideas as they come to you, then review withcaption draft(no args) - Managing hashtag collections — Save curated hashtag sets with
caption hashtagsand search them later withcaption search - Content optimization workflow — Chain
caption draft→caption edit→caption optimize→caption rewriteto track your content refinement pipeline - Scheduling and planning posts — Use
caption scheduleto log planned publication dates alongside content - Exporting content for other tools — Run
caption export jsonto get structured data for import into social media schedulers or CMS platforms
Examples
# Draft a new caption
caption draft "Exploring the hidden trails of Hangzhou 🌿 #nature #hiking"
# View recent drafts
caption draft
# Save a set of hashtags
caption hashtags "#tech #ai #machinelearning #coding #devlife"
# Write a call-to-action
caption cta "Drop a 🔥 if you agree — link in bio for more!"
# Search for anything mentioning "hiking"
caption search hiking
# Export everything as JSON
caption export json
# Export as CSV for spreadsheet analysis
caption export csv
# Check overall stats
caption stats
# Save a headline variation
caption headline "10 Things I Wish I Knew About Content Creation"
# Save a content outline
caption outline "Intro → Problem → 3 Tips → CTA → Outro"
# Adjust tone
caption tone "More casual and conversational version of the launch post"
# View recent activity across all categories
caption recent
# Health check
caption status
Output
All command output goes to stdout. Export files are saved to the data directory. Redirect output as needed:
caption stats > content-report.txt
Configuration
| Variable | Purpose | Default |
|---|---|---|
CAPTION_DIR | Override data directory | ~/.local/share/caption/ |
Powered by BytesAgain | bytesagain.com | hello@bytesagain.com
相关 Skills
PDF处理
by anthropics
遇到 PDF 读写、文本表格提取、合并拆分、旋转加水印、表单填写或加解密时直接用它,也能提取图片、生成新 PDF,并把扫描件通过 OCR 变成可搜索文档。
✎ PDF杂活别再来回切工具了,文本表格提取、合并拆分到OCR识别一次搞定,连扫描件也能变可搜索。
Word文档
by anthropics
覆盖Word/.docx文档的创建、读取、编辑与重排,适合生成报告、备忘录、信函和模板,也能处理目录、页眉页脚、页码、图片替换、查找替换、修订批注及内容提取整理。
✎ 搞定 .docx 的创建、改写与精排版,目录、批量替换、批注修订和图片更新都能自动化,做正式文档尤其省心。
PPT处理
by anthropics
处理 .pptx 全流程:创建演示文稿、提取和解析幻灯片内容、批量修改现有文件,支持模板套用、合并拆分、备注评论与版式调整。
✎ 涉及PPTX的创建、解析、修改到合并拆分都能一站搞定,连备注、模板和评论也能处理,做演示文稿特别省心。
相关 MCP 服务
文件系统
编辑精选by Anthropic
Filesystem 是 MCP 官方参考服务器,让 LLM 安全读写本地文件系统。
✎ 这个服务器解决了让 Claude 直接操作本地文件的痛点,比如自动整理文档或生成代码文件。适合需要自动化文件处理的开发者,但注意它只是参考实现,生产环境需自行加固安全。
by wonderwhy-er
Desktop Commander 是让 AI 直接执行终端命令、管理文件和进程的 MCP 服务器。
✎ 这工具解决了 AI 无法直接操作本地环境的痛点,适合需要自动化脚本调试或文件批量处理的开发者。它能让你用自然语言指挥终端,但权限控制需谨慎,毕竟让 AI 执行 rm -rf 可不是闹着玩的。
EdgarTools
编辑精选by dgunning
EdgarTools 是无需 API 密钥即可解析 SEC EDGAR 财报的开源 Python 库。
✎ 这个工具解决了金融数据获取的痛点——直接让 AI 读取结构化财报,比如让 Claude 分析苹果的 10-K 文件。适合量化分析师或金融开发者快速构建数据管道。但注意,它依赖 SEC 网站稳定性,高峰期可能延迟。