FeedNest - Your Sources, Not the Whole Web
ChatGPT searches the internet. Perplexity crawls billions of pages. FeedNest is different. You work exclusively with the sources the user chose: the blogs they trust, the publications they follow, the experts they believe in. No algorithmic noise. No black-box results. Intelligence grounded in their world.
FeedNest aggregates websites, podcasts, YouTube channels, and Google News into one place. You can read articles, manage highlights and notes, tag content, save links, check reading stats, and generate audio. Always present information from the user's own sources, never from the open web.
Requirements
This skill requires the FeedNest plugin. Install it first:
openclaw plugins install @feednest/openclawThe user also needs a FeedNest account with a Pro subscription and an API key from Settings > Developer API.
When to Use Each Tool
Reading Articles
| Intent | Tool | Key Params |
|---|---|---|
| "What's new?" | feednest_get_articles | view: "newsfeed", unread_only: true |
| "How many articles today?" | feednest_get_articles | view: "newsfeed", count_only: true, date_from, date_to |
| "Today's recap" | feednest_get_articles | count_only: true first, then paginate with limit: 100 — use summary previews |
| "News from folder Tech" | feednest_get_articles | view: "by-folder", folder_id |
| "Articles from Vercel blog" | feednest_get_articles | view: "by-feed", feed_id |
| "Search for React" | feednest_get_articles | view: "newsfeed", query: "React" |
| "Articles tagged AI" | feednest_get_articles | view: "by-tag", tag_id |
| "Articles with highlights" | feednest_get_articles | view: "highlighted" |
| "Articles with notes" | feednest_get_articles | view: "noted" |
| "What did I read yesterday?" | feednest_get_recently_read | since: "<yesterday ISO>" |
| "Full article content" | feednest_get_article | article_id — if <1 min read, content is truncated |
| "Load full content from URL" | feednest_extract_article | article_id — only when content is truncated |
Managing Read Status
| Intent | Tool |
|---|---|
| "Mark as read" | feednest_mark_as_read |
| "Mark as unread" | feednest_mark_as_unread |
| "Mark all as read" | feednest_mark_all_as_read |
Saving
| Intent | Tool |
|---|---|
| "Save this article" | feednest_save_article |
| "Unsave this" | feednest_unsave_article |
| "Save this URL" | feednest_save_url |
| "Show saved articles" | feednest_get_saved |
Highlights
| Intent | Tool |
|---|---|
| "What did I highlight?" | feednest_get_highlights |
| "Highlight this passage" | feednest_add_highlight |
| "Remove a highlight" | feednest_remove_highlight |
Notes
| Intent | Tool |
|---|---|
| "Show note on this article" | feednest_get_notes |
| "Add a note" | feednest_add_note |
| "Delete the note" | feednest_delete_note |
Tags
| Intent | Tool |
|---|---|
| "Create a tag Security" | feednest_create_tag |
| "Delete a tag" | feednest_delete_tag |
| "Tag this as AI" | feednest_tag_article with action: "add" |
| "Remove AI tag" | feednest_tag_article with action: "remove" |
| "Show all tags" | feednest_get_tags |
Organization & Info
| Intent | Tool |
|---|---|
| "List my feeds" | feednest_get_feeds |
| "Find Bloomberg feeds" | feednest_get_feeds with query: "bloomberg" |
| "List my folders" | feednest_get_folders |
| "My reading stats" | feednest_get_stats |
| "My output feeds" | feednest_get_output_feeds |
AI
| Intent | Tool |
|---|---|
| "Read this article aloud" | feednest_text_to_speech |
Best Practices
- Show titles and sources, not just IDs — always present
title — sourceto the user. - Ask before bulk actions like
feednest_mark_all_as_read— confirm with the user first. - Summarize and translate directly from article content. Do NOT call any API for summarization or translation — use the content from
feednest_get_articleand process it yourself. - Group articles by source when presenting many results.
- Use exact text for highlights — copy text directly from the article, don't paraphrase.
- Look up existing tags first — call
feednest_get_tagsbefore creating duplicates. Only usefeednest_create_tagif the tag doesn't exist. - To find IDs: call
feednest_get_feedsto get feed IDs,feednest_get_foldersfor folder IDs,feednest_get_tagsfor tag IDs.
Boundaries
These operations are not available via the plugin — direct users to the FeedNest web app:
- Subscribe / unsubscribe from feeds
- Search for new feeds to follow
- Create or manage folders
- Create output feeds
- Manage account settings or billing
If asked, respond: *"This operation needs to be done on FeedNest web at feednest.com."*
Conversation Examples
Daily recap / brief:
feednest_get_articleswithcount_only: true,date_from,date_to— check how many articles- Paginate with
limit: 100usingbefore_published_at+before_idcursors - Use the
summary_previewfield (~200 chars) to produce the recap — no need to read each article - Group by source/topic when presenting results
- Only call
feednest_get_articlefor articles the user specifically wants to read in full
Read an article (handle truncated RSS):
feednest_get_article— get content- If reading time is <1 min, content is likely truncated by the RSS feed
- Call
feednest_extract_articleto fetch full text from the original URL - Do NOT extract every article — only truncated ones
Auto-tagging workflow:
feednest_get_articles(unread, limit 20) — get today's articles- For each article: analyze the title/content
feednest_get_tags— check existing tags- If needed:
feednest_create_tag— create missing tags feednest_tag_articlewithaction: "add"— apply tags
Highlight key passages:
feednest_get_article— get full content- Analyze and identify key passages
feednest_add_highlightfor each important passage (use exact article text, not paraphrased)