SKILL·98BE4C

download-wechat-channels

hekmon8
更新于 4 days ago
6 次查看
在 GitHub 上查看
其他general

关于

该Skill用于安装和操作wx_channels_download工具,以保存用户已授权的微信视频号内容。它支持两种下载路径:直接使用视频URL和密钥的CLI模式,或通过本地HTTPS代理进行交互式抓取。该工具适用于macOS、Windows和Linux系统,并能处理下载器诊断及二进制更新等操作。

快速安装

Claude Code

推荐
主要方式
npx skills add hekmon8/wx-channels-download-skill -a claude-code
插件命令备选方式
/plugin add https://github.com/hekmon8/wx-channels-download-skill
Git 克隆备选方式
git clone https://github.com/hekmon8/wx-channels-download-skill.git ~/.claude/skills/download-wechat-channels

在 Claude Code 中复制并粘贴此命令以安装该技能

技能文档

Download WeChat Channels

Use the upstream wx_channels_download binary without copying its implementation. Only download videos the user owns or is authorized to save.

Choose the path

  • If the user has a direct video URL and optional decryption key, use the direct CLI path. This does not require changing the system proxy.
  • Otherwise use the interactive WeChat path. It starts a local HTTPS interception proxy, may install a root certificate, and may modify the system proxy.
  • Do not claim that an ordinary WeChat share URL is sufficient. Upstream share-link parsing needs extra service credentials and is not the default local workflow.

Install or update

Run:

python3 scripts/wx_channels.py install
python3 scripts/wx_channels.py doctor

The installer downloads the matching asset from the latest GitHub release, verifies it against the release checksum file, and stores it outside the skill repository. It prefers the non-UPX safe Windows build. Do not redistribute the downloaded binary.

Direct URL download

Confirm the URL came from a source the user is authorized to download. Then run:

python3 scripts/wx_channels.py download --url "$VIDEO_URL" --filename "video.mp4"

Add --key NUMBER only when a decryption key was captured with the URL. The upstream command writes to the user's Downloads directory.

Interactive WeChat download

Before running this path, explain that it can install a local root certificate and change the system proxy. Obtain explicit user authorization for those system changes.

After authorization:

python3 scripts/wx_channels.py start --allow-security-changes

Keep the process running. Ask the user to open the authorized video in the desktop WeChat client, wait for playback, pause it, and use the injected download button. Report the download directory printed by the process.

Stop with Ctrl+C. Verify that the upstream process restored the previous proxy settings; if network access is abnormal, run its uninstall command:

python3 scripts/wx_channels.py uninstall --allow-security-changes

Do not run start or uninstall without explicit authorization. Do not expose captured URLs, cookies, tokens, keys, certificates, or log contents in chat or commits.

Troubleshoot

Run doctor first. Read references/upstream.md for architecture, supported platforms, ports, conflicts, and license constraints. Prefer upstream documentation and releases over remembered flags.

GitHub 仓库

hekmon8/wx-channels-download-skill
路径: skills/download-wechat-channels
FAQ

常见问题

什么是 download-wechat-channels Skill?

download-wechat-channels 是一个 Claude Skill,作者为 hekmon8。Skill 将 Claude 按需加载的说明和资源打包,让 Claude 无需额外提示即可执行与 download-wechat-channels 相关的任务。

如何安装 download-wechat-channels?

使用本页的安装命令:将 download-wechat-channels 作为插件添加到 Claude Code,或将其仓库克隆到 skills 目录,然后重启 Claude 以加载该 Skill。

download-wechat-channels 属于哪个分类?

download-wechat-channels 属于其他分类。

download-wechat-channels 可以免费使用吗?

可以。download-wechat-channels 已收录在 AIMCP,可免费安装。

相关推荐技能

llamaguard
其他

LlamaGuard是Meta推出的7-8B参数内容审核模型,专门用于过滤LLM的输入和输出内容。它能检测六大安全风险类别(暴力/仇恨、性内容、武器、违禁品、自残、犯罪计划),准确率达94-95%。开发者可通过HuggingFace、vLLM或Sagemaker快速部署,并能与NeMo Guardrails集成实现自动化安全防护。

查看技能
cost-optimization
其他

这个Claude Skill帮助开发者优化云成本,通过资源调整、标记策略和预留实例来降低AWS、Azure和GCP的开支。它适用于减少云支出、分析基础设施成本或实施成本治理策略的场景。关键功能包括提供成本可视化、资源规模调整指导和定价模型优化建议。

查看技能
sports-betting-analyzer
其他

该Skill为开发者提供体育博彩数据分析工具,可分析盘口、大小球和特殊投注,识别价值投注机会。它整合历史数据和情景统计,生成包含时间戳的结构化Markdown报告。适用于需要快速获取博彩市场洞察的娱乐或教育类应用开发。

查看技能
quantizing-models-bitsandbytes
其他

这个Skill使用bitsandbytes库量化大语言模型,能在GPU内存有限时通过8位或4位量化减少50-75%内存占用,同时保持精度损失最小。它支持INT8、NF4、FP4等多种量化格式,可与HuggingFace Transformers无缝集成,适用于需要部署更大模型或加速推理的场景。还提供QLoRA训练和8位优化器支持,让开发者能轻松实现高效模型压缩。

查看技能