About
A personal knowledge wiki for AI work.
Wenlan is your personal knowledge wiki for AI work. Save useful decisions with their sources, then find and build on them across your AI tools.
Project status
v0.21.7
macOS, Linux, Windows
Apache-2.0 + AGPL-3.0
Built by Qi-Xuan Lu
Why Wenlan exists
AI work often leaves useful decisions, debugging lessons, and project details buried in old chats. Wenlan gives that knowledge a home you can return to.
Save what matters, add trusted sources, and shape them into pages you can inspect and find again across your AI tools.
Design principles
Local-first
Wenlan keeps its knowledge on your machine by default. The model you configure can run locally or through a cloud provider; cloud sync and telemetry are optional.
Human-readable
Wenlan keeps readable Markdown files in local git. A local database supports search, and you can inspect the source-linked files yourself.
Session rhythm
Wenlan helps carry useful context from one AI work session to the next: start with pages, save decisions, and prepare a handoff when you finish.
Deliberate distillation
Run /wenlan distill when a topic is ready to become a source-linked page. With optional model setup, Wenlan can also organize pages automatically and add richer links between ideas.
Built by Qi-Xuan Lu
Wenlan is built and maintained by Qi-Xuan Lu (GitHub @7xuanlu). Background in AI infrastructure, knowledge graphs, and local-first systems.
Wenlan is designed so agents can add to it and people can inspect it. It keeps source-linked pages and session logs in readable files, while one local service makes them available to connected AI tools.
Project channels: GitHub Issues for bugs and feature requests, SECURITY.md for vulnerabilities, and the Wenlan release notes for changes.
Current status
Wenlan v0.21.7 includes a notarized macOS Apple Silicon DMG, a Windows x64 desktop installer, and headless runtime packages for macOS, Linux (x86_64 and aarch64; glibc), and Windows (x86_64). The daemon, CLI, MCP server, Claude Code plugin, and Codex plugin use Apache-2.0. The desktop app crate uses AGPL-3.0-only.