Who smallweb is for#
Independent readers tired of algorithmic content feeds
Browse personal blogs and genuine writing without engagement bait or SEO noise. Kagi Small Web surfaces posts from individual creators writing for their own audiences, organized by topic so you can find content in areas you care about, with a "Next Post" button for serendipitous discovery.
Skip if:
Skip if you already have a well-curated RSS reading list and follow specific writers consistently. The main value is discovery; once you know which blogs to follow, a standard RSS reader handles the ongoing workflow more efficiently.
Personal bloggers seeking discovery beyond search
Submit your personal blog's RSS feed to the index via GitHub pull request, alongside two other blog submissions. Inclusion means your posts appear on the Kagi Small Web site and in Kagi search results, reaching readers specifically seeking authentic personal content rather than SEO-optimized articles.
Skip if:
Skip if your blog is multi-author, carries ads or affiliate links, has no posts in the last 12 months, or publishes LLM-generated content. Each of those is a disqualifying condition under the current inclusion rules.
Developers building tools for the indie web corpus
The RSS feed (4,320+ entries) and OPML file are public resources for building applications on top of the small web dataset. The MIT license and plain-text feed format make them straightforward to consume programmatically for analysis, alternative interfaces, or content aggregation projects.
Skip if:
Skip if you need a corpus of mainstream or high-volume publications. The dataset is specifically personal blogs and small independent channels, which excludes major media outlets and large YouTubers by design.
Independent YouTubers with fewer than 100,000 subscribers
YouTube channels in the small web index must post no more than twice weekly and have fewer than 100,000 subscribers. Qualifying channels get their videos surfaced alongside personal blogs, reaching an audience specifically looking for independent, non-commercial video content rather than algorithmic recommendations.
Skip if:
Skip if your channel exceeds 100,000 subscribers or posts more than twice weekly. Both are hard disqualifiers under the inclusion rules, with no exception process currently documented.
The problem it solves#
The mainstream web has become difficult to read. Search engines surface SEO-optimized articles that all say the same thing. Social media feeds prioritize engagement bait over genuine writing. Discovering a personal blog written by a real person with something original to say requires either knowing exactly where to look or getting lucky through someone else's recommendation.
Feed aggregators like Feedly and Inoreader help once you know which feeds to follow, but they do not solve the discovery problem. You have to find the blog first. The small web is the corner of the internet filled with personal sites, independent channels, and webcomics from individual creators, and it is increasingly invisible to anyone who relies on algorithmic surfaces for their reading material.
How it solves it#
Community-maintained RSS list
The discovery index is built from three plain text files on GitHub: personal blogs (smallweb.txt), YouTube channels (smallyt.txt), and independent webcomics (smallcomic.txt). Anyone can submit a new feed via a GitHub pull request, with the requirement to also add at least two other sites in the same commit. This keeps the list community-owned rather than editorially gated.
Topic-based browsing across 20+ categories
Posts are organized into topic categories including AI, Programming, Science, DIY & Making, Health & Fitness, Photography, Gaming, and Life & Personal. Each category shows a live count of active entries. Readers can browse within a topic or jump randomly to any recent post using the "Next Post" button.
Seven-day rolling feed
Only posts from the last seven days appear on the site, keeping the content fresh and preventing stale results from dominating. Posts older than a week are not displayed through the small web interface, though they remain discoverable through Kagi search. The full RSS feed aggregates all 4,320+ indexed feeds in one subscription.
Kagi search integration
Blogs in the small web index also surface in Kagi search results when queries are relevant to their content. This means small web posts reach Kagi users even without the discovery site, extending reach for independent creators who pass the curation criteria without requiring any separate submission.
Strict inclusion criteria for authentic content
The index has explicit rules against auto-generated, LLM-generated, or spam content. Multi-author blogs, commercial sites, and blogs with ads or undisclosed affiliate links are excluded. YouTube channels must have fewer than 100,000 subscribers and post no more than twice weekly. These rules are enforced by community review on pull requests.
Strengths and trade-offs#
Strengths
- MIT licensed with publicly available application codeThe full application powering the Kagi Small Web website is MIT licensed and hosted on GitHub under kagisearch/smallweb. The feed lists are plain text files editable through standard GitHub pull requests, keeping the content layer independent of any single host or proprietary format.
- No algorithmic ranking or engagement optimizationPosts appear in simple chronological order. There is no engagement-based ranking, no promoted content, and no advertising. The curation is editorial: a community-maintained allowlist rather than an algorithm. This means the feed reflects what people actually wrote, not what drives clicks.
- Active development with a strong contributor communityWith 1,664 GitHub stars and 728 forks, the project has attracted a contributor base that actively submits personal blogs and channels. The repo received its most recent push in August 2026, indicating ongoing maintenance. The community requirement to submit two other blogs per contribution creates a peer-discovery flywheel.
- RSS and OPML export prevent reader lock-inThe full small web index is available as an RSS feed (4,320+ entries) and an OPML file for bulk import. Any RSS reader can subscribe to the aggregated feed or import the OPML to follow individual sources directly. Readers who want more control are not locked into the Kagi interface.
Trade-offs
- -English-only content at this timeThe index currently accepts only English-language content. Non-English personal blogs, YouTube channels, and webcomics are excluded regardless of quality or subject matter. International readers and creators writing in other languages have no equivalent curated path through this project.
- -Seven-day freshness gate limits archive browsingThe website shows only posts from the last seven days. Older posts from indexed blogs are not browsable through the small web interface, though they may appear in Kagi search results. Readers who want access to a creator's full archive must visit that individual blog directly.
- -No deployment documentation for self-hostingWhile the application is MIT licensed and the source code is public, the README does not include instructions for running your own instance. The app/ directory contains the code that powers the site, but self-hosters must read it directly to understand the setup requirements. There is no Docker image or quickstart guide.
smallweb vs alternatives#
Kagi Small Web vs Feedly
Feedly and Kagi Small Web both aggregate content from RSS feeds, but they serve different purposes. Feedly is a commercial RSS management tool for following sources you already know. Kagi Small Web is a discovery surface for finding sources you have never heard of.
| Feature | Kagi Small Web | Feedly |
|---|---|---|
| License | MIT (open source) | Proprietary |
| Cost | Free | Free tier limited; paid plans from $6/month |
| Content scope | Personal blogs, indie YouTube, webcomics only | Any RSS feed |
| Curation model | Community-vetted allowlist | User-configured subscriptions |
| Self-hosting | Possible (no deployment docs) | No |
Kagi Small Web is the better choice when the goal is discovery: finding personal blogs and independent creators you would not encounter through search or social media. Feedly is the better choice once you have identified which sources to follow and want a managed reading experience with folder organization, read state tracking, search, and team sharing features.
Kagi Small Web vs Inoreader
Inoreader is a more feature-complete RSS reader than Feedly, with automation rules, filters, and content monitoring. Kagi Small Web does not try to compete with this feature set.
| Feature | Kagi Small Web | Inoreader |
|---|---|---|
| License | MIT (open source) | Proprietary |
| Cost | Free | Free tier; paid from $7.99/month |
| Content scope | Curated small web only | Any RSS feed |
| Filtering and rules | None | Advanced |
| Self-hosting | Possible (no deployment docs) | No |
Inoreader is a power tool for managing a large reading list across many sources. Kagi Small Web is a front door to the personal web for readers who do not yet have a reading list. The two are complementary: many readers could use Kagi Small Web for discovery and then add interesting blogs to Inoreader for ongoing reading.
What it's built on#
- Languages
- JavaScriptPython
- Frameworks
- Flask
FAQ#
Is Kagi Small Web free to use?
Yes. The Kagi Small Web site at kagi.com/smallweb is free to browse with no account required. The RSS feed and OPML file are publicly accessible. The application code is MIT licensed and available on GitHub at no cost. Access to small web posts within Kagi search results does require a Kagi subscription, since Kagi is a paid search engine, but the discovery site itself has no paywall.
How do I add my blog to the Kagi Small Web index?
Submit your blog's RSS or Atom feed via a GitHub pull request to the smallweb.txt file in the kagisearch/smallweb repository. The rules require adding at least two other blogs you do not own in the same commit. Content must be in English, the blog must be personal (no multi-author publications), it must have a post within the last 12 months, and the site must not carry ads, undisclosed affiliate links, or popup overlays.
How does Kagi Small Web differ from Feedly or Inoreader?
Feedly and Inoreader are general-purpose RSS readers where you subscribe to any feed you choose and manage your own reading list. Kagi Small Web is a curated index of vetted personal blogs only, focused on discovery rather than feed management. You browse what is already in the index rather than building your own subscription list. The service is also free and open source, while Feedly and Inoreader charge for full-featured access.
Can I self-host my own Kagi Small Web instance?
Self-hosting is possible in principle: the application is MIT licensed and the full source code is available on GitHub under the app/ directory. However, the README does not include deployment instructions, so you would need to read the source code directly to understand the setup requirements. The feed list files (smallweb.txt, smallyt.txt, smallcomic.txt) are the community data layer and can be used independently of the web application.
What types of content are excluded from the small web index?
The index excludes multi-author blogs, commercial blogs with advertising or affiliate links, AI-generated or LLM-generated content, sites without a post in the last 12 months, non-English content, YouTube channels with more than 100,000 subscribers, and YouTube channels posting more than twice weekly. Comics must be independently created and RSS feeds must show the full comic in the feed. Commercially syndicated comics are also excluded.
Similar open-source tools#
Craft CMS
Flexible developer-friendly CMS for content teams
rybbit
Cookieless web analytics: GDPR compliant, 18 KB script
diagram-design
Editorial diagrams for Claude Code, no Figma or Mermaid required
agency-agents
Expert AI agent personalities for every workflow
open-seo
Open source alternative to Semrush and Ahrefs
listmonk
High-performance newsletter and mailing list manager