360 Secure SkillHub
360 安全技能中心

ClawVision Plus

ClawVision Plus 是 ClawVision 的配套插件,不独立生成摘要卡片;它读取主 Skill 产出的 JSON 与 HTML,导出每个标签页一页的多页 PDF、生成 1200×630 OG 预览图,并可在用户确认后通过 Telegram Bot 发送图片和摘要文案。需要 Python 3.10+ 及 Pillow、reportlab、playwright、python-tele…

monaxamo
monaxamo
热度 13.30
最近更新 2026-08-28 14:30:08
分类:效率依赖:代码执行 / 读写文件 / 联网浏览搜索 / 第三方 API应用场景:文件转换 / 图像生成 / 社交媒体

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 clawvision-plus
Metadata
name
clawvision-plus
description
Companion plugin for ClawVision: add PDF, OG images, and Telegram sharing.
metadata
{"version":"1.0.0","author":"Maximius","tags":["clawvision","plugin","companion","pdf","og-image","telegram","export","social","openclaw"],"homepage":"https://github.com/monaxamo/clawvision","license":"MIT","icon":"clawvision_demo_en.png"}
allowed-tools
["write","read","exec","message"]
user-invocable
true

ClawVision Plus — companion plugin for ClawVision

This is not a standalone skill. It extends ClawVision with three extra outputs for summaries it already generated.

Adds

  1. PDF export — multi-page PDF, one page per ClawVision tab (Main / Format / Built / Next).
  2. OG image — 1200×630 social-preview PNG.
  3. Telegram sharing — send the OG image + summary caption to a Telegram chat or channel.

When to use

Only after the main clawvision skill has generated the HTML card. Good triggers:

  • "Export this ClawVision summary as PDF."
  • "Generate an OG image for this ClawVision card."
  • "Send this ClawVision summary to Telegram."
  • "Run ClawVision Plus on the summary we just made."

Workflow

  1. Use clawvision to generate the summary card (HTML + JSON).
  2. Run clawvision-plus/scripts/extend_visual.py on the same JSON and output directory.
  3. Get the PDF, OG image, and/or Telegram delivery.

Usage

# 1. Generate summary with the main ClawVision skill
python clawvision/scripts/generate_visual.py \
  --summary summary.json \
  --output ./out \
  --png --md --pptx \
  --lang en

# 2. Run the companion plugin
python clawvision-plus/scripts/extend_visual.py \
  --summary summary.json \
  --output ./out \
  --pdf --og \
  --telegram \
  --telegram-chat-id "@your_channel" \
  --telegram-bot-token "YOUR_BOT_TOKEN"

Parameters

  • --summary — path to ClawVision summary JSON.
  • --output — directory containing the rendered HTML card.
  • --slug — optional slug override.
  • --pdf — export HTML to a multi-page PDF, one page per tab.
  • --og — generate 1200×630 OG image.
  • --telegram — send OG image + caption to Telegram.
  • --telegram-chat-id — Telegram chat/channel ID.
  • --telegram-bot-token — Telegram bot token.
  • --telegram-caption — override the generated caption.

Requirements

  • Python 3.10+
  • playwright (already required by ClawVision)
  • Pillow
  • reportlab (for multi-page PDF)
  • python-telegram-bot (only for Telegram sharing)

Safety

  • Do not send summaries to Telegram without explicit user confirmation.
  • Never share Telegram bot tokens in chat or logs.
  • The plugin does not upload raw session transcripts.