Token导航 LogoToken导航

Agent Skill · 研究检索

streaming-buddy(流媒体好友)

streaming-buddy 用于记录任务执行中的错误、用户纠正、经验和能力缺口,适合在 OpenClaw 中希望让 Agent 持续沉淀问题、修正和最佳实践时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

streaming-buddy 是研究检索类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为接触敏感数据;许可证为 MIT-0,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 1,851。

总安装 74,635周安装 3,079GitHub Stars 1下载 24,386核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT-0
版本
2.0.0

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:streaming-buddy(流媒体好友)
来源仓库:https://github.com/udiedrichsen/streaming-buddy
安装命令:
openclaw skills install streaming-buddy
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install streaming-buddy

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

具有学习偏好的个人流媒体助理。跟踪您正在观看的内容,了解您的品味,并根据您的服务、心情和偏好建议接下来观看的内容。当被问及电影、电视节目、流媒体服务、观看内容或跟踪观看进度时使用。

触发因素:/stream、“我应该看什么”、“推荐一些东西”、提及 Netflix/Prime/Disney+/Apple TV+、询问剧集/季节/剧集、基于情绪的请求(例如“令人兴奋的东西”)。

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 OpenClaw;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 openclaw skills install streaming-buddy;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要接触敏感数据,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

streaming-buddy 怎么安装?

在终端执行 openclaw skills install streaming-buddy。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

streaming-buddy 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

streaming-buddy 需要什么权限?

权限级别是接触敏感数据。该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

streaming-buddy 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

streaming-buddy 用的人多吗?

公开数据是累计安装 74,635 次,近一周安装 3,079 次,下载 24,386 次,GitHub 1 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

streaming-buddy 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

值得选的地方和要先想清楚的地方

值得选的地方

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。

先想清楚的地方

  • 需要接触敏感数据,装到常用环境前要先评估影响面。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 OpenClaw,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 环境里有大量密钥、凭证,不方便给第三方 Skill 读取的人。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
streaming-buddy

流媒体好友 · 本页

74,6351接触敏感数据MIT-0有提醒

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

name
streaming-buddy
version
2.0.0
description
Personal streaming assistant with learning preferences. Tracks what you're watching, learns your taste, and suggests what to watch next based on your services, mood, and preferences. Use when asked about movies, TV shows, streaming services, what to watch, or tracking viewing progress. Triggers: /stream, 'what should I watch', 'recommend something', mentioning Netflix/Prime/Disney+/Apple TV+, asking about series/seasons/episodes, mood-based requests like 'something exciting'.
author
clawdbot
license
MIT
metadata
clawdbot
emoji
📺
triggers
["/stream"]
requires
bins
["jq", "curl"]
env
["TMDB_API_KEY"]
tags
["streaming", "movies", "tv-shows", "recommendations", "entertainment", "learning", "preferences"]

Streaming Buddy 📺

Personal streaming assistant that learns your taste, tracks your watching habits, and suggests what to watch next.

Features

  • Search & Info: Find movies/TV shows with TMDB data
  • Watch Tracking: Track what you're currently watching with progress
  • Learning System: Learns your preferences from likes/dislikes/ratings
  • Smart Recommendations: Personalized suggestions based on your taste
  • Mood-Based Search: Find content by mood (exciting, relaxing, scary, etc.)
  • Availability Check: Shows which of your services has the content
  • Match Explanation: Explains why a title matches your preferences

Commands

CommandAction
/streamShow status with all commands
/stream search <title>Search for movies/TV shows
`/stream info <id> [tv\movie]`Detailed info + availability
`/stream watch <id> [tv\movie]`Start tracking a title
/stream progress S01E05Update progress on current show
/stream done [1-5]Mark as finished + rate (auto-learns)
/stream like [id]Mark as liked → learns preferences
/stream dislike [id]Mark as disliked → learns preferences
`/stream suggest [service] [tv\movie]`Personalized recommendations
/stream mood <mood>Search by mood
/stream surpriseRandom recommendation
/stream why <id>Explain why this matches you
/stream watchlistShow watchlist
/stream watchlist add <id>Add to watchlist
/stream historyView watch history
/stream profileShow your taste profile
/stream servicesManage streaming services
/stream services add <name>Add a service
/stream services remove <name>Remove a service

Mood Options

MoodGenres
excitingAction, Thriller, Sci-Fi, Adventure
relaxingComedy, Animation, Family, Documentary
thoughtfulDrama, Mystery, History
scaryHorror, Thriller
romanticRomance, Drama
funnyComedy, Animation

Supported Services

  • netflix, amazon-prime, disney-plus, apple-tv-plus
  • youtube-premium, wow, paramount-plus, crunchyroll
  • joyn, rtl, magenta, mubi

Learning System

The skill learns your preferences from:

  1. Ratings: When you finish with /stream done [1-5]:

- Rating 4-5: Adds genres/themes/actors to "liked" - Rating 1-2: Adds genres to "avoided"

  1. Explicit Feedback: /stream like and /stream dislike:

- Extracts genres, themes, actors, directors - Updates preference weights

  1. Preference Profile includes:

- Genre preferences (weighted scores) - Liked/disliked themes - Favorite actors & directors - Custom mood mappings

Handler Usage

# Core commands
handler.sh status $WORKSPACE
handler.sh search "severance" $WORKSPACE
handler.sh info 95396 tv $WORKSPACE
handler.sh watch 95396 tv $WORKSPACE
handler.sh progress S01E05 $WORKSPACE
handler.sh done 5 "Great show!" $WORKSPACE

# Learning commands
handler.sh like $WORKSPACE                    # Like current watching
handler.sh like 12345 movie $WORKSPACE        # Like specific title
handler.sh dislike $WORKSPACE
handler.sh why 95396 tv $WORKSPACE
handler.sh profile $WORKSPACE

# Recommendation commands
handler.sh suggest $WORKSPACE                 # All services, all types
handler.sh suggest prime movie $WORKSPACE     # Prime movies only
handler.sh mood exciting $WORKSPACE
handler.sh mood relaxing tv $WORKSPACE
handler.sh surprise $WORKSPACE

# List commands
handler.sh watchlist list $WORKSPACE
handler.sh watchlist add 12345 tv $WORKSPACE
handler.sh history $WORKSPACE

# Service management
handler.sh services list $WORKSPACE
handler.sh services add netflix $WORKSPACE
handler.sh services remove netflix $WORKSPACE

Data Files

All data stored in $WORKSPACE/memory/streaming-buddy/:

FilePurpose
config.jsonTMDB API key, region, language
profile.jsonUser profile metadata
services.jsonActive streaming services
preferences.jsonLearned taste preferences
watching.jsonCurrently watching
watchlist.jsonWant to watch list
history.jsonWatched + ratings
cache/*.jsonAPI response cache (24h)

Setup

  1. Get TMDB API key: https://www.themoviedb.org/settings/api
  2. Store in memory/streaming-buddy/config.json:
   {
     "tmdbApiKey": "your_api_key",
     "region": "DE",
     "language": "de-DE"
   }
  1. Run /stream setup to configure services

Conversation Examples

Mood-based search:

User: I want something exciting tonight
Bot: 🎬 Exciting picks for you:
     1. Reacher S3 (Prime) ⭐8.5
     2. Jack Ryan (Prime) ⭐8.1
     ...

Learning from feedback:

User: /stream done 5
Bot: ✅ Severance marked as done (⭐5)
     📚 Learned: +Drama, +Mystery, +Sci-Fi
     Actors: Adam Scott, Britt Lower saved to favorites

Explaining recommendations:

User: /stream why 95396
Bot: 🎯 Why Severance matches you:
     ✓ Genre "Drama" (you like this, +2)
     ✓ Genre "Mystery" (you like this, +2)
     ✓ Theme "office" in your preferences
     ✓ With Adam Scott (your favorite)
     Similar to: Fallout ⭐5

Language Support

  • Language detected from config.json (language: "de-DE" or "en")
  • All output adapts to configured language
  • Commands work in any language

Requirements

  • jq (JSON processor)
  • curl (HTTP client)
  • bash 4.0+
  • TMDB API key (free)

References

平台分布

OpenClaw

71.36%
按下载量换算17,402

安全审计

VirusTotal

通过

ClawScan

可疑

Static analysis

未展示

权限和风险

接触敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在提醒或未通过项,安装前请自行复核。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills