智能收音机

agent-radio

by Artem Pisarev

Control internet radio: play streams, search stations, manage favorites, volume, stop/pause/next, and inspect current playback. Uses mpv with audio device auto-detection and ffplay fallback. Settings persist between runs.

4.5kAI 与智能体未扫描2026年3月30日

安装

- id: brew-mpv

文档

Agent Radio Skill

Internet radio player for OpenClaw with station search, favorites, volume control, and playback management.

When to Use

Use when user wants to:

  • Play an internet radio stream by URL
  • Search for stations by name from the live Radio Browser directory
  • Play a random built-in station when no target is provided
  • Manage favorite stations (add, remove, list)
  • Control volume, pause/resume, stop, skip to next
  • Check what is currently playing
  • Have a persistent radio experience across sessions

Tools

This skill provides the following CLI commands in scripts/:

play <url|station_name> [volume]

Play a stream by URL or by name from favorites or the built-in list.

If no target is provided, the skill chooses a random station from stations.json. Optional volume is 0-100 and defaults to the saved setting.

Examples:

  • play https://stream.zeno.fm/0r0xa792kwzuv
  • play BBC Radio 1
  • play "Jazz 24 (KNKX)" 60

stop

Stop playback immediately.

pause

Pause or resume the current playback process.

next

Skip to the next favorite station in a cycle.

volume [level]

Show the current volume or set a new default volume from 0-100.

Examples:

  • volume
  • volume 75

now

Show the current station URL, volume, PID, and playback status.

favorite add <name> <url>

Add a station to favorites.

favorite remove <name>

Remove a station from favorites.

list

List all favorite and built-in stations.

find <query> [number]

Search the Radio Browser directory. When number is provided, play that result immediately.

Examples:

  • find jazz
  • find "lofi hip hop" 1

Configuration

Preferences are stored in: {baseDir}/preferences.json

json
{
  "last_station": "",
  "volume": 80,
  "favorites": [
    {"name": "BBC Radio 1", "url": "http://..."},
    {"name": "Jazz 24 (KNKX)", "url": "https://..."}
  ],
  "audio_device": "auto",
  "current_pid": null,
  "paused": false
}

Audio device auto-detection:

  • macOS: coreaudio/BuiltInSpeakerDevice
  • Linux: alsa/default
  • Windows: directsound/default

Override locations when needed:

  • AGENT_RADIO_BASE_DIR
  • AGENT_RADIO_PREF_FILE
  • AGENT_RADIO_STATIONS_FILE

Implementation Notes

  • Primary player: mpv --no-video --audio-device=<device> --volume=<vol> --cache=yes <url>
  • Fallback player: ffplay -nodisp -autoexit <url>
  • Playback runs in the background and stores PID plus paused state in preferences.
  • On play, any existing playback is stopped before starting the next stream.
  • Station lookup is case-insensitive across favorites and built-in stations.
  • next cycles favorite URLs instead of depending on station-name matching.

Error Handling

  • If both mpv and ffplay are missing, return install guidance.
  • If dependencies like jq or curl are missing, fail early with a clear message.
  • If a stored PID is stale, clear it automatically.
  • If station search returns no matches, explain how to retry with another query or a direct URL.

Examples for Users

text
/radio play https://stream.zeno.fm/0r0xa792kwzuv
/radio play "BBC Radio 1"
/radio volume 70
/radio now
/radio find jazz
/radio find "lofi hip hop" 1
/radio favorite add "Lounge Jazz" https://jazz.stream
/radio list
/radio next
/radio stop

Future Enhancements

  • Built-in station directory with genres and countries
  • Recording streams to file
  • Equalizer presets
  • Web interface control

相关 Skills

Claude接口

by anthropics

Universal
热门

面向接入 Claude API、Anthropic SDK 或 Agent SDK 的开发场景,自动识别项目语言并给出对应示例与默认配置,快速搭建 LLM 应用。

想把Claude能力接进应用或智能体,用claude-api上手快、兼容Anthropic与Agent SDK,集成路径清晰又省心

AI 与智能体
未扫描164.6k

RAG架构师

by alirezarezvani

Universal
热门

聚焦生产级RAG系统设计与优化,覆盖文档切块、检索链路、索引构建、召回评估等关键环节,适合搭建可扩展、高准确率的知识库问答与检索增强应用。

面向RAG落地,把知识库、向量检索和生成链路系统串联起来,做架构设计时更清晰,也更少踩坑。

AI 与智能体
未扫描23.3k

多智能体架构

by alirezarezvani

Universal
热门

聚焦多智能体系统架构设计,梳理 Supervisor、Swarm、分层和 Pipeline 等模式,覆盖角色定义、通信协作与性能评估,适合规划稳健可扩展的 AI agent 编排方案。

帮你系统解决多智能体应用的架构设计与协同编排难题,适合构建复杂 AI 工作流,成熟度高、社区认可也很亮眼。

AI 与智能体
未扫描23.3k

相关 MCP 服务

知识图谱记忆

编辑精选

by Anthropic

热门

Memory 是一个基于本地知识图谱的持久化记忆系统,让 AI 记住长期上下文。

帮 AI 和智能体补上“记不住”的短板,用本地知识图谱沉淀长期上下文,连续对话更聪明,数据也更可控。

AI 与智能体
89.0k

顺序思维

编辑精选

by Anthropic

热门

Sequential Thinking 是让 AI 通过动态思维链解决复杂问题的参考服务器。

这个服务器展示了如何让 Claude 像人类一样逐步推理,适合开发者学习 MCP 的思维链实现。但注意它只是个参考示例,别指望直接用在生产环境里。

AI 与智能体
88.1k

by deusdata

热门

持久化的代码库知识图谱,可跨会话保留上下文,在 session 重启或上下文压缩后仍能继续使用。

专治 AI 编程助手“会话失忆”,把代码库沉淀为持久知识图谱,重启或压缩上下文后也能无缝续上开发状态。

AI 与智能体
26.9k

评论