← 返回 Skills 市场
35
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install feeds-digest
功能描述
Multi-Source-RSS-Digest (YouTube, MS Tech Community, GitHub) mit Datums- und Topic-Filter und optionaler LLM-Zusammenfassung.
使用说明 (SKILL.md)
feeds-digest
Sammelt Updates aus RSS/Atom-Feeds, filtert nach Zeitraum und Themen, gibt einen Markdown-Digest aus.
Schnellstart
bash scripts/install.sh # einmalig
feeds-digest --since 7d # Standard-Digest
feeds-digest --since 3d --topics bc,fabric # gefiltert
feeds-digest --llm # mit LLM-Summary
feeds-digest --output report.md # in Datei
feeds-digest --test # Feed-Erreichbarkeit prüfen
Quellen
- YouTube-Kanäle (RSS)
- Microsoft Tech Community (RSS)
- GitHub Releases (Atom)
- Generische RSS/Atom-Feeds
Siehe
README.md— ausführliche Dokureferences/youtube-channel-discovery.md— Channel-IDs findenreferences/ms-tc-categories.md— MS TC Kategorienconfig/config.example.yaml— Config-Vorlage
安全使用建议
Install only if you are comfortable with this skill fetching configured feeds and writing local cache/history files. Do not use --llm with private or sensitive feeds unless you are comfortable sending the digest content to the configured provider; use Ollama for local-only summarization. Consider pinning dependencies or using a lockfile in controlled environments.
能力标签
能力评估
Purpose & Capability
The skill collects RSS/Atom feed items from configured YouTube, Microsoft, GitHub, and generic RSS sources, filters them, and produces Markdown or JSON output; these capabilities match its stated purpose.
Instruction Scope
The README and SKILL.md disclose feed fetching, file output, optional LLM summaries, API-key use, and manual cron examples, but the LLM section could be clearer that digest content is sent to the selected external provider when Perplexity or OpenAI is used.
Install Mechanism
The installer creates user config/cache directories, copies an example config, and installs Python dependencies with pip --user; it does not install cron jobs automatically, though dependencies are loosely versioned rather than pinned.
Credentials
Network access, configured feed URLs, and optional provider API keys are proportionate for a feed aggregation and summarization tool; no hidden credential harvesting, destructive actions, or unrelated data access was found.
Persistence & Privilege
The skill stores seen GUIDs and a local daily history of feed titles/links under ~/.cache/feeds-digest for deduplication/history; this is purpose-aligned and partly documented, but retention is not bounded.
如何使用
- 确保已安装 OpenClaw(本地或 Docker 部署)
- 在对话框中输入安装命令:
/install feeds-digest - 安装完成后,直接呼叫该 Skill 的名称或使用
/feeds-digest触发 - 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
Initial release of feeds-digest:
- Aggregates updates from multiple RSS/Atom sources (YouTube, Microsoft Tech Community, GitHub, and generic feeds).
- Allows filtering by date range and topics.
- Outputs a Markdown digest.
- Optional LLM-based summary generation.
- Includes commands to test feed accessibility and output digests to file.
- Configuration examples and additional references provided.
元数据
常见问题
Feeds Digest 是什么?
Multi-Source-RSS-Digest (YouTube, MS Tech Community, GitHub) mit Datums- und Topic-Filter und optionaler LLM-Zusammenfassung. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 35 次。
如何安装 Feeds Digest?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install feeds-digest」即可一键安装,无需额外配置。
Feeds Digest 是免费的吗?
是的,Feeds Digest 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
Feeds Digest 支持哪些平台?
Feeds Digest 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。
谁开发了 Feeds Digest?
由 jcbuer(@jcbuer)开发并维护,当前版本 v1.0.0。
推荐 Skills