Most "free backlink" lists check whether a platform lets you publish and whether the link is dofollow. Both can be true on a page search engines are told to ignore. rentry.co was the best candidate we tested: anonymous publishing, no account, and outbound links rendered with an empty rel attribute. Every page also carries <meta name="robots" content="noindex">.
So the order of checks matters, cheapest first: can you publish without an account, does the page load, is it indexable, did your link survive rendering, and is it followed. The third one kills most ideas in a single request.
Telegraph (telegra.ph)Passes link valueno account · 2026-09-11
index, follow. The API issues its own token, so there is no signup at all. Body links carry rel="nofollow" but the author link does not, so each page gives one followed link.
graph.orgPasses link valueno account · 2026-09-14
Telegram serves every Telegraph page on this second domain at the same path, index, follow, with the same followed author link. Nothing to publish: the page appears by itself.
te.legra.phPasses link valueno account · 2026-09-14
A third Telegram domain serving the identical page, also index, follow, also with the followed author link. One publish, three domains.
Dev.toPasses link valuefree account · 2026-09-11
Indexable, links followed, and the REST API accepts canonical_url, so the syndicated copy points home instead of competing with the original. The only human step is creating the account once.
pkg.go.devNofollow - traffic onlyno account · 2026-09-14
Indexable, and it creates a page for any public Go module automatically with no account anywhere. Every outbound link we checked was nofollow, so this is discovery, not ranking weight.
libraries.ioNofollow - traffic onlyno account · 2026-09-14
Auto-indexes packages from many registries, indexable, homepage link is nofollow.
pypi.orgNofollow - traffic onlyfree account · 2026-09-14
Indexable, project links are nofollow.
LinkedIn (posts, articles, company page)Nofollow - traffic onlyfree account · 2026-09-14
Googlebot and Bingbot are explicitly allowed in robots.txt while everyone else gets Disallow: /, so pages are indexable. But every outbound link is rewritten through the lnkd.in redirector AND carries rel="nofollow" - checked across a company page, including the website field. The only followed links on the page point at other LinkedIn domains. Zero ranking value passed to your site, by design.
GitHub profile and repositoriesNofollow - traffic onlyfree account · 2026-09-11
Indexable and high authority, but links are nofollow. Worth having for people, not for rankings.
rentry.conoindex - passes nothingno account · 2026-09-11
Publishes anonymously and renders outbound links with rel="" - fully followed, body links included. Then serves every page with <meta name="robots" content="noindex">. The best-looking candidate we tested and worth nothing.
anotepad.comnoindex - passes nothingno account · 2026-09-11
noindex on every note, and the URL is not rendered as a link at all.
npmjs.com package pagesnoindex - passes nothingfree account · 2026-09-14
Package pages carry <meta name="robots" content="noindex,nofollow">. Publishing a package for the link is wasted effort.
Write.as (anonymous)Cannot publishno account · 2026-09-11
Accepts the POST with 201 and then serves the result at /contentisblocked. Their filter drops anonymous posts containing outbound links.
dpaste, paste.ee, controlc, justpaste.itCannot publishblocked · 2026-09-11
No anonymous write: 405, application key required, 403, and 404 respectively.
AI tool directories (aitools.fyi, Futurepedia and similar)Paid submissionpaid · 2026-09-11
aitools.fyi submission is a $37 paid boost. Others are paid tiers or refuse non-browser requests. Also the wrong category for a services business.
The set of platforms that are indexable, give a followed link, and need no account at all is essentially Telegraph and its two mirror domains. That is not a gap in the research. Anywhere a link carries weight, the platform defends itself against exactly this kind of automation, and it should.
Which leaves two honest routes. One free account per platform, after which publishing can run on its own through an API - that is what we do with Dev.to. Or content people choose to link to, which is slower and is the only one that compounds.
If a page or a list tells you otherwise, run the indexability check on it before you spend an afternoon. That takes one request.