Tags

Tags organize related articles across feeds and categories. RSSMonster stores tags per user, displays them on articles, and surfaces frequently used tags in the sidebar for quick filtering.

Where Tags Come From

Tags can be added automatically during a feed crawl or manually from an article.

Manual Tags

You can add and manage tags on individual articles, including articles already in your library. These changes apply to the selected article; removing a tag from it does not remove the same label from other articles.

Add tags to an article

1. Open the article’s three-dot Article actions menu and choose Add tags.
2. Select existing labels from the suggestions or use Search tags to find a label by name.
3. To create a label, select Create new tag, enter a name, and select Create. You can also type a new name into the search field and choose Create “name” when offered.
4. Select Add tags to save the selected labels to the article.

Creating a label selects it in the dialog; it is saved to the article only when you submit your changes. Cancel closes the dialog without saving.

Open the article’s … menu and choose Add tags to select existing labels or create a new one inline. Existing article tags remain selected and cannot be removed in this workflow. Current tags appear immediately. Suggestions combine up to ten names from the current sidebar’s usage snapshot (Most used in this view) with ten alphabetical defaults, without repeating selected labels. Search looks up matching names on the server after a 250 ms pause, returning up to twenty results with selected matches first. Clearing search restores the small suggestion set. Create “name” selects a new label without closing the dialog. The dialog never loads the complete tag collection.

There is no Recently used section: crawl processing can recreate tag rows, so their creation timestamps do not reliably represent when you last used a tag.

Manage an article’s tags

1. Open the article’s three-dot Article actions menu and choose Manage tags.
2. Review Current tags at the top of the dialog. Select the × on a tag chip or uncheck its matching search result to remove it from the selection.
3. Under Add more tags, search for and select existing labels, choose from Most used in this view or More tags, or use Create new tag.
4. Select Save changes to apply additions and removals together, or Cancel to leave the article unchanged.

The screenshot shows Verstappen and Videos selected on one article. Remove either chip to stop assigning that label to this article, or select additional labels before saving. Removing an automatically assigned tag can be temporary: later crawl processing may add it again. Manually added tags survive automatic tag reconciliation.

Saved tags appear immediately on the article, including compact and mobile metadata. They use the same search, Smart Folder, and sidebar navigation as automatically assigned tags. If a multi-request save is interrupted, the dialog keeps your selections and retries only the remaining changes.

Publisher Tags and Categories

RSS and Atom providers can include categories, labels, or tags in an article. RSSMonster extracts those values and carries them into the article’s tags automatically. You do not need to recreate labels that the publisher already provides.

Publisher values are normalized before they are stored. Empty and duplicate values are removed so the same label is not displayed more than once on an article.

Feed Tags

You can configure tags on an RSSMonster feed. Those feed tags are applied to new articles saved from that feed, which is useful for stable labels such as a publisher, project, team, or broad subject area.

Generated Tags

When AI enrichment is configured for a feed, RSSMonster can derive a small set of tags from the article content. These are combined with categories supplied by the publisher.

Rule-Based Tags

Actions can assign your own tag when an incoming article matches a regular expression. Rules are useful when publishers do not provide the label you need or when you want one consistent tag across several sources.

For example, if you follow Nintendo and Zelda news, create an Automation action with:

• Name: Nintendo and Zelda
• Type: Assign tag
• Tag value: nintendo
• Regular Expression: [Nn]intendo|[Zz]elda

The rule checks each incoming article’s title, HTML and plain-text body, description, and URL. If any one of those fields matches, RSSMonster assigns the nintendo tag. This example uses character classes so it matches both capitalized and lowercase forms.

Rules run while articles are processed during a crawl. They are intended for incoming articles and should not be treated as a retroactive search over the existing library.

Rule-based tags and Smart Folder queries solve different problems. An Automation rule uses a JavaScript regular expression to attach a persistent tag during article processing. A Smart Folder uses a search expression to build a dynamic view without modifying the article.

To reject matching entries before they are stored or enriched, configure a feed item filter instead.

Finding Tagged Articles

Tags appear as clickable labels on the articles themselves. Rule-assigned tags have their own visual treatment so they can be distinguished from other tag sources.

RSSMonster also shows the most frequently used tags for the active article status in the sidebar. Each sidebar tag includes the number of matching articles in that status.

Select a tag on an article or in the sidebar to show other articles carrying the same tag. The current status, sorting, grouping, and view mode remain in effect, so an Unread view continues to show only unread articles with that tag. Select the active sidebar tag again to clear the tag filter.

You can also use a tag directly in Search or a Smart Folder query:

tag:nintendo

Combine it with other expressions to narrow the result further:

tag:nintendo unread:true @lastweek sort:recommended

Quote multiword tag names in search expressions, for example tag:"read later". See the Search Guide for every supported expression.

Article Tagging API

Authenticated API clients can assign manual tags to existing articles. There is no separate tag catalogue: a new label is created when it is first assigned to an article, and remains available while at least one assignment exists.

Operation	API
Default suggestions	GET /api/tags?scope=all&limit=10
Search available names	GET /api/tags?search=security&limit=20
Read current assignments	GET /api/articles/:articleId (article.tags)
Add existing or new labels	POST /api/articles/:articleId/tags with {"tags":["security","project"]}
Remove one assignment	DELETE /api/articles/:articleId/tags/:tagId

The available-name response is { "tags": [{ "name": "security" }], "hasMore": false }. Names are distinct and alphabetically ordered across the user’s articles, regardless of reading status. limit defaults to 100 and accepts 1–100; offset defaults to zero. Advance the offset by the limit while hasMore is true. Supplying search selects this same name catalogue (with or without scope=all). Search is normalized using the existing trim/lowercase convention and matches literal substrings; % and _ are not wildcards. An exact match is ordered first, followed by alphabetical matches. Search defaults to twenty results and accepts limits of 1–50. Queries longer than 255 characters after normalization are rejected. Both tag and article ownership are enforced in the same limited SQL query. Existing user/name indexes are reused; no popularity aggregation runs on keystrokes. Without scope=all or search, GET /api/tags retains its status-scoped Top Tags behavior.

Add requests accept 1–100 string names. Names are trimmed and lowercased using the same conventions as automatically assigned tags; empty names and names longer than 255 characters are rejected. Repeated assignments are idempotent. New assignments have tagType: "manual". An existing assignment keeps its original provenance and is not converted from a rule, feed, or provider tag.

Both mutation endpoints return { "tags": [...] } with the article’s current assignments, each containing id, name, and tagType. The removal ID identifies one assignment on that article, not a label shared across the library. Removal does not affect another article’s assignments, even when the names match. An explicitly removed automatic tag can return during later crawl processing. Manual tags survive automatic tag reconciliation.

Mutations require an owned, visible canonical article. Malformed IDs or invalid payloads return 400; nonexistent, foreign-owned, or unavailable articles and assignments return 404. Authentication uses the existing API session middleware. Changes are immediately available to tag search and Smart Folder queries. Multiword labels work in search and Smart Folders using a quoted tag expression, for example tag:"read later". The Smart Folder editor preserves the complete tag name and generates the required quotes.