← 返回 Skills 市场
odysseus0

Feed Digest

作者 George Zhang · GitHub ↗ · v1.0.0
cross-platform ✓ 安全检测通过
1087
总下载
0
收藏
4
当前安装
1
版本数
在 OpenClaw 中安装
/install feed-digest
功能描述
Agentic feed digest using the feed CLI. Fetch, triage, and summarize RSS/Atom/JSON feeds to surface high-signal posts. Use when: (1) reading feeds or catching up on news, (2) user asks for a digest, roundup, or summary of recent posts, (3) user asks what's new or interesting today, (4) user mentions feed, RSS, blogs, or subscriptions.
使用说明 (SKILL.md)

RSS Digest

Surface what's worth reading from your feeds. Requires feed CLI (brew install odysseus0/tap/feed).

Workflow

  1. Fetchfeed fetch to pull latest entries.
  2. Scanfeed get entries --limit 50 for recent unread (title, feed, date, summary).
  3. Triage — Pick 5-10 high-signal posts. Prioritize: AI progress, systems engineering, developer tools, anything surprising or contrarian.
  4. Readfeed get entry \x3Cid> for each pick (full post as Markdown).
  5. Synthesize — For each post: title, source, 2-3 sentence summary of why it matters. Group by theme if natural clusters emerge.
  6. Mark readfeed update entries --read \x3Cid1> \x3Cid2> ... to mark triaged entries as read.

Commands

feed fetch                              # pull latest from all feeds
feed get entries --limit N              # list unread entries (table)
feed get entries --feed \x3Cid> --limit N  # filter by feed
feed get entry \x3Cid>                     # read full post (Markdown)
feed search "\x3Cquery>"                   # full-text search
feed update entries --read \x3Cid> ...     # batch mark read
feed get feeds                          # list feeds with unread counts
feed get stats                          # database stats

Notes

  • Default output is table — most token-efficient for scanning. Avoid -o json.
  • feed get entry \x3Cid> returns Markdown — read this for the actual post content.
  • Filter by feed if too many entries: --feed \x3Cfeed_id>.
安全使用建议
This skill is coherent with its description, but before installing: (1) inspect the Homebrew tap and GitHub repo (odysseus0/feed) to ensure you trust the author, (2) prefer installing in a controlled environment or sandbox if you have concerns, (3) be aware the tool will fetch remote feeds and modify your local feed database (it will mark items as read), and (4) if you subscribe to private authenticated feeds, understand the CLI may use stored credentials—verify how those are stored/used. If you don't trust the tap/repo, don't install the formula; consider running the CLI manually after reviewing its source.
功能分析
Type: OpenClaw Skill Name: feed-digest Version: 1.0.0 The skill bundle provides instructions for an AI agent to use the `feed` CLI tool for digesting RSS/Atom/JSON feeds. All commands and workflow steps described in `SKILL.md` are directly related to the stated purpose of fetching, triaging, reading, and summarizing feed entries. There is no evidence of data exfiltration, malicious execution, persistence mechanisms, obfuscation, or prompt injection with harmful intent. The dependency on an external `feed` CLI tool (from `odysseus0/tap/feed` or `github.com/odysseus0/feed`) is clearly stated and its usage is aligned with the skill's function, without any indication of malicious exploitation of the tool itself.
能力评估
Purpose & Capability
Name/description (feed digest) match the declared requirement of a single 'feed' CLI binary and the SKILL.md commands. No unrelated credentials, files, or binaries are requested.
Instruction Scope
Instructions are narrowly scoped to running the 'feed' CLI: fetch entries, read specific entries, triage and summarize, and mark entries read. This will read feed content and modify the local feed DB/state (marking items read); the skill does not request unrelated system files or environment variables. Be aware the agent will run network fetches and change local feed state.
Install Mechanism
Install options are a third-party Homebrew tap (odysseus0/tap/feed) and a Go package from github.com/odysseus0/feed. Both are traceable but not an official core/homebrew formula — installing will build/run code from that author. Verify the tap/repo before installation because those install sources can execute arbitrary code on your machine.
Credentials
No environment variables or external credentials are requested by the skill definition. Note that the 'feed' CLI itself may access user feed configuration or credentials for private feeds (not declared here). The lack of requested credentials is consistent with a public-feed reader skill.
Persistence & Privilege
Skill is not always-enabled and does not request system-wide privileges beyond installing/running the 'feed' binary. Runtime actions include marking items read in the local feed database (normal for this purpose). Installation will place a binary via Homebrew or Go tooling (expected behavior).
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install feed-digest
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /feed-digest 触发
  4. 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 feed-digest skill. - Enables fetching, triaging, and summarizing RSS/Atom/JSON feeds for high-signal posts using the feed CLI. - Provides clear workflow steps: fetch, scan, triage, read, synthesize, and mark entries as read. - Includes detailed command reference and setup instructions for Homebrew and Go installations. - Designed for quick digests, news roundups, and surfacing noteworthy content from user subscriptions.
元数据
Slug feed-digest
版本 1.0.0
许可证
累计安装 4
当前安装数 4
历史版本数 1
常见问题

Feed Digest 是什么?

Agentic feed digest using the feed CLI. Fetch, triage, and summarize RSS/Atom/JSON feeds to surface high-signal posts. Use when: (1) reading feeds or catching up on news, (2) user asks for a digest, roundup, or summary of recent posts, (3) user asks what's new or interesting today, (4) user mentions feed, RSS, blogs, or subscriptions. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 1087 次。

如何安装 Feed Digest?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install feed-digest」即可一键安装,无需额外配置。

Feed Digest 是免费的吗?

是的,Feed Digest 完全免费(开源免费),可自由下载、安装和使用。

Feed Digest 支持哪些平台?

Feed 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 Feed Digest?

由 George Zhang(@odysseus0)开发并维护,当前版本 v1.0.0。

💬 留言讨论