跳至正文
OpenAgentSkill
浏览已开放 · 目录记录日期 2026-11-21 · 自动更新和账户服务暂时关闭。
其他技能

rss-reader

Parse and monitor RSS and Atom feeds using the feedparser Python library

mrveiss · 已有 AI 审核记录

安装与使用

运行命令前先阅读来源与依赖说明。复制命令不会自动安装或执行。

npx skills add mrveiss/AutoBot-AI --skill rss-reader

技能说明


name: rss-reader version: 1.0.0 description: Parse and monitor RSS and Atom feeds using the feedparser Python library author: mrveiss category: internet tools:

  • parse_feed
  • list_entries
  • get_entry_content
  • check_feed_info triggers:
  • read RSS feed
  • parse RSS
  • Atom feed
  • subscribe to feed
  • latest news from
  • check blog updates
  • feed URL
  • RSS URL
  • feedparser tags:
  • rss
  • atom
  • feed
  • feedparser
  • news
  • monitoring
  • content-aggregation

Capability Tiers

┌─────────────────────────────────────────────────────────────┐
│ rss-reader                                                    │
├─────────────────────────────────────────────────────────────┤
│ STANDALONE   (always works, zero config)                     │
│   • Parse any public RSS/Atom feed with feedparser           │
│     (tolerant of malformed XML, no auth)                     │
├─────────────────────────────────────────────────────────────┤
│ SUPERCHARGED (unlocks when you connect a tool)               │
│   • atoma parser         → typed Atom objects on strict feeds │
│   • Custom User-Agent /  → feeds that reject the default      │
│     Accept header           request                          │
└─────────────────────────────────────────────────────────────┘

STANDALONE — always works, zero config

  • Parse any public RSS or Atom feed with feedparser; both formats are detected automatically and malformed XML is tolerated.
  • List titles, extract entries as JSON, read full entry content, and inspect feed metadata — all without authentication.

SUPERCHARGED — unlocks per connected tool

Connect thisUnlocks
atoma parserStrongly-typed Atom entry objects for strict Atom feeds where feedparser degrades.
Custom User-Agent / Accept header handlerFeeds that reject the default request (403 / empty body) become readable.

These are graceful fallbacks — the skill defaults to feedparser and only needs the extras for the minority of feeds that require them.

Output Contract

Feed reads return this fixed shape:

## Feed: <feed title>

**URL:** <feed_url>   **Type:** rss | atom   **Entries:** <n>
**Parser:** feedparser | atoma

| # | Title | Published | Link |
|---|-------|-----------|------|
| 1 | ...   | ...       | ...  |

Single-entry content mode returns the raw entry body (HTML or plain text) under a ## <entry title> heading.

When to Use

Use this skill when an agent needs to read news, blog posts, or any content published via an RSS or Atom feed. feedparser handles both feed formats automatically, tolerates malformed XML, and requires no authentication for public feeds.

Workflow

List entry titles from a feed

python3 -c "
import feedparser
d = feedparser.parse('{feed_url}')
for e in d.entries[:10]:
    print(e.title)
"

Get entries as JSON with metadata

python3 -c "
import feedparser, json
d = feedparser.parse('{feed_url}')
entries = [
    {
        'title': e.title,
        'link': e.link,
        'published': e.get('published', ''),
        'summary': e.get('summary', '')[:200],
    }
    for e in d.entries[:10]
]
print(json.dumps(entries, indent=2))
"

Check feed metadata (title, type, entry count)

python3 -c "
import feedparser
d = feedparser.parse('{feed_url}')
print('Feed title:', d.feed.get('title', 'N/A'))
print('Feed version:', d.version)
print('Entry count:', len(d.entries))
print('Bozo (parse error):', d.bozo)
"

Get full content of the first entry

python3 -c "
import feedparser
d = feedparser.parse('{feed_url}')
e = d.entries[0]
content = e.get('content', [{}])[0].get('value', e.get('summary', 'No content'))
print(content)
"

Fallback — atoma (typed objects, Atom feeds only)

python3 -c "
import atoma, requests
feed = atoma.parse_atom_bytes(requests.get('{feed_url}').content)
for e in feed.entries[:5]:
    print(e.title.value)
"

Output Format

  • List mode: One entry title per line; suitable for quick enumeration.
  • JSON mode: Array of entry objects with title, link, published, and summary fields.
  • Content mode: Raw HTML or plain text body of the selected entry; pass through a markdown converter if needed.

Entry Fields Reference

FieldDescription
titleEntry headline
linkURL to the full article
summary / descriptionEntry excerpt or full text
published / updatedDate string (not always present)
authorAuthor name
tagsList of category/tag objects
contentFull content list (Atom feeds only)

Parameters

ParameterTypeRequiredDescription
feed_urlstringyesFully-qualified RSS or Atom feed URL
limitintnoMaximum number of entries to return (default 10)
fieldstringnoSpecific field to extract (e.g., title, link)

Limitations

  • feedparser does not execute JavaScript; feeds served via JS redirects will parse empty.
  • Entry dates are not guaranteed — use e.get('published', e.get('updated', '')) for safe access.
  • Very large feeds (thousands of entries) may be slow; slice with d.entries[:n].
  • Some feeds require an Accept header or specific User-Agent; pass a custom handler if the default request is rejected.
  • feedparser does not validate feed signatures or SSL certificates beyond the system trust store.

Fallback Instructions

If fe

阅读完整来源 ↗