Flowise, the open-source low-code builder for LLM apps and agents, is shutting down — the team has frozen development, will archive the GitHub repository on 10 August 2026, and marks the project end-of-life on 31 August 2026

Flowise is one of the tools that made “build an LLM app without writing much code” real for a lot of teams: a visual, node-based canvas where you wire a model to a retriever, some memory, and a set of tools, and get a chatbot, a RAG pipeline, or an agent out the other end — self-hosted from the open-source project or run through its cloud. Its maintainers have now announced they are winding down. In their words, “we’ve decided to wind down our operations for Flowise.”

Unlike a lot of shutdown notices, this one comes with a clear calendar. Feature development stopped at a 29 July 2026 announcement and code freeze. On 10 August 2026 the GitHub repository is archived — it becomes read-only, issues and pull requests are locked, and the npm packages and Docker images are marked deprecated. 31 August 2026 is the stated end-of-life, when the core team’s presence around the project concludes. The code itself does not disappear: it stays public and, being Apache-2.0 licensed, remains free to use and fork indefinitely.

For a self-hosted deployment, nothing breaks on any of those dates — your instance keeps running exactly as it did. What changes is that after 10 August the project is frozen. There will be no more releases, no more fixes, and no security patches, so any future vulnerability in Flowise or in the large dependency tree underneath it becomes something you own and patch yourself, on a fork, with no upstream to pull from. That is the quiet kind of dependency risk: it does not cause an outage today, it causes an audit finding and an unpatched CVE some months from now. The maintainers are candid about this and encourage teams to “fork the repo to maintain their own internal updates or community-led forks.”

The cloud side is the sharper deadline. The page does not spell out what happens to hosted accounts, data, or deployments after end-of-life, so anyone relying on Flowise Cloud should treat 31 August as the date to have exported everything they need and moved their workloads, rather than waiting to find out. There is no named replacement product and no automated migration, which means the realistic options are: fork and self-maintain, adopt a community fork if a serious one emerges, or rebuild the flow on another visual LLM builder or a code-first framework — none of which import a Flowise graph directly. The one thing worth doing before 10 August, whatever you choose, is the cheap one: fork the repository now, while the full history and issue tracker are still open, so the source of the tool you depend on stays in your hands.