Science1 distinct publisher2 min readUpdated
shinyapps.io stops taking new apps at the end of 2026 and existing content is moved automatically in early 2027. Deployment pipelines built for it need editing before then.
The Scientist · Science desk
Compiled by The ScientistSomething wrong?How this is made
The break is narrow and mechanical, which is why it tends to sit untouched until a deploy fails. The pattern published on r-bloggers three years ago was a commit hook: push to the repository, and a GitHub Actions job bundled the Shiny source and sent it to shinyapps.io [6]. Most of that machinery survives, because the deployment methods that worked against shinyapps.io still work against Posit Connect Cloud [7]. What changes is the paperwork at each end. The Actions route now begins by obtaining a Content ID [8]. The Git-backed route wants a manifest.json committed alongside the code, which in R is usually one call to `rsconnect::writeManifest()` [9]. Neither is difficult. Both are edits to a workflow file nobody has opened since it was written, on content Posit will move for you in early 2027 whether the edit happened or not [2].
The post's own recommendation is to stop deploying from Actions and let Connect Cloud watch the branch, on the grounds that the deployment work no longer consumes your GitHub Actions allowance [10]. That saves compute minutes, and it also moves where trust sits, since the platform is granted access to the GitHub account [9]. The same post lists the cases where that trade does not close, and they are not exotic for anyone with a review process. You give up deploying only after earlier pipeline checks have passed [11]. You may be barred from granting a third-party tool access to an account that holds sensitive code elsewhere [12]. On the free tier with a private repository, the option is not on the table [13].
The migration therefore sorts teams by how their deployment was built rather than by what their app does. Anyone who shipped by clicking Deploy in RStudio and thinking no further about packages or containers [21] has little to do beyond noting that a free tier survived [4]. Anyone with a gated pipeline has a small rewrite with a deadline attached. On price the direction is not uniform: paid plans sit above their nearest shinyapps.io equivalents [5], while the custom domain that used to require the $349 per month Professional tier now appears on the $59 per month Enhanced tier with an SSL certificate [15][16].
The schedule is the part worth writing down. According to the r-bloggers account, Posit gave roughly eleven months between the January 2026 announcement and the end-of-2026 close to new apps [18], then named a part of a year rather than a date for the automatic migration [2]. A pipeline that breaks during that migration breaks at a moment somebody else picked.
Follow any of these and your For You feed starts watching them — no settings page required.
Ranked by verification strength, evidence, and original report placement.
In January 2026 Posit announced that shinyapps.io would close to new apps at the end of 2026, with all users moving to Posit Connect Cloud.
Existing content on shinyapps.io will continue to work before automatically migrating to Posit Connect Cloud in early 2027.
Posit Connect Cloud is not restricted to Shiny: it also supports Streamlit, Bokeh and Jupyter Notebooks, and all plans allow unlimited hosting for rendered Quarto and R Markdown documents.
Posit Connect Cloud adds the ability to set secret variables, regenerate content on a schedule, higher maximum compute limits, and SSL certificates with custom domains.
Paid tiers of Posit Connect Cloud are priced higher than their nearest shinyapps.io equivalents.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Specific but single-sourced
The factual core is unusually concrete for a single article — named dates, named tiers, exact prices, and reproducible deployment steps down to rsconnect::writeManifest() and the Content ID location. But everything rests on one practitioner post aggregated via r-bloggers.com, with no link to or quotation from Posit's own announcement and no second publisher to corroborate the timeline or the tier pricing.
No usage evidence
Nothing in the supplied material quantifies adoption on either side of the migration: no app counts, user counts, migration progress, or disclosed Posit Connect Cloud usage. The two observations available are a vendor sunset announcement and a tier price change, which describe supply-side decisions rather than measured uptake, so no adoption value can be assigned without guessing.
Slightly overstated upside
The article is mostly sober instruction and it does flag the downsides — higher paid tiers, third-party repo access, free-tier plus private-repo limits. The mild overstatement is in framing: Connect Cloud 'stacks more features and convenience on top' and custom-domain-only users are 'perhaps the biggest winners', which reads the forced migration as an upgrade, while the piece supplies no evidence that the automatic early-2027 migration will be clean or that most teams come out ahead on cost.
Practitioner content marketing, disclosed
The post is a consultancy-style tutorial republished through r-bloggers.com; it promotes the author's own prior article and its own recommended workflow, and it benefits from being the reference guide for a migration the whole R community must perform. That is a visible but ordinary content-marketing incentive, not a vendor payment — the author has no disclosed commercial stake in Posit, and openly names situations where Posit's own Git-backed deployment should be avoided.
Moderate
Confidence is capped by single-publisher sourcing and the absence of the primary Posit announcement, but supported by the internal specificity and falsifiability of the claims: dates, prices and commands that any R practitioner could check within minutes. The deployment mechanics deserve higher confidence than the timeline and pricing figures.
build
A file-copy Allure adapter for Katalon, and the history IDs that make retries useful1 distinct publisher
build
A NetworkPolicy in another repo broke invoicing while every dashboard reported success1 distinct publisher
build
A cleanup commit deleted the sanitizer. Five days later a scanner cashed it in.1 distinct publisher
build
Flux moves GitOps' source of truth into registries you own, and mirroring becomes the prerequisite1 distinct publisher
Distinct publishers with included, body-backed reporting in this cluster.
1 article · August 21, 2026