Beautiful architecture diagrams that support real-time updates can now be generated with a single click using AI!
We’re sure you’ve been frustrated by the need to constantly adjust and stretch your canvas during manual drawing, and by how easily your images become outdated after code updates.
While generating with Mermaid is convenient, more complex diagrams may result in scattered arrows that still require manual adjustments.
Now, an open-source project called Archify, which has made it to GitHub Trending, has condensed this into one sentence:
Analyze this code repository and generate a high-level runtime architecture diagram.
Then, an HTML-formatted architecture diagram is generated, allowing you to search for nodes, trace call paths, and switch themes.
And it works with any model—whether you're using Claude Code, Codex, Cursor, or any other agent.
As of publication, Archify has garnered numerous positive reviews with over 100,000 likes on X, accumulated 1.9K forks and 31.4K stars, and surged to the top of GitHub’s trending list.
Everyone couldn't help but exclaim:
Thank you, Archify—I’ve saved at least four or five follow-up meetings.
Surprisingly, the developer of this popular AI agent, "Wu Feng Wu Qing," did not graduate from a prestigious university.
He completed a junior college program and was admitted to Chongqing University of Posts and Telecommunications’ Software Engineering program with the top score on the transfer exam.
After Archify went viral, he playfully remarked with a touch of sadness on his X account:
I’ve worked as a security guard, learned car repair, worked in a factory, and even made it through multiple rounds of interviews at major companies like ByteDance’s Feishu and Yuanfudao, always thinking the job offer was just within reach—only to be rejected during background checks when they discovered I attended a junior college. I had no choice but to walk away in disappointment. Every year after graduation, I’d face the same situation: the interview went smoothly, but as soon as I mentioned I was a “junior college to bachelor’s” graduate, the HR response would go silent. (Sigh)
He candidly admitted that after one offer after another fell through, he suffered, questioned himself, felt lost, and even immediately ran five kilometers downstairs.
But he never gave up.
Later, encouraged by Manus COO Cz Chen, he chose to use Archify to prove to everyone that:
Coming from humble beginnings is not shameful!
Turn code into a system map in one step.
Archify is an Agent Skill designed for AI agents such as Claude Code, Codex, Cursor, and OpenCode.
You can provide a system description, or let Archify analyze your code repository directly.
After understanding the code, Archify first generates a structured JSON IR, then performs schema, layout, and rendering checks, and finally outputs a standalone HTML file.
The process generally works as follows:
Code or text description → Archify understanding → Generate JSON IR → Rule validation → Render HTML/SVG → Interaction and export.
It can currently generate five types of technical diagrams: architecture diagrams, workflow diagrams, sequence diagrams, data flow diagrams, and lifecycle diagrams.
Moreover, the images it generates aren’t just for viewing—you can search for nodes, examine upstream and downstream modules, trace paths between two nodes, freely switch between light and dark themes, and export in PNG, SVG, WebM formats, as well as shareable cards.
If you're not satisfied with the result, you can also ask Archify to make changes anytime, like this:
Add Redis.
Move the authentication module to the left.
Highlight the fallback path after cache invalidation.
But the question arises: tools like Mermaid and D2 for generating diagrams from text have existed for a long time—what makes Archify unique?
The answer is that it’s not just about generating images from text—it’s about embedding drawing capabilities directly into the programming agent’s workflow, enabling the agent to understand code, generate GIF charts, and continuously modify them.
In comparison, Mermaid and Graphify are primarily responsible for rendering existing text descriptions into images, and they may not even produce results as visually appealing as Archify.
It's not just a prettier Mermaid
What sets Archify apart is that it integrates diagramming tools directly into the workflow of its programming agents.
You don't need to learn drawing syntax from scratch—just describe your technical intent.
The work of understanding code, organizing components, and generating structured data can be delegated to an AI Agent, while Archify handles verification and rendering.
This division of labor also aims to address one of the issues with AI-generated architecture diagrams, namely
The model can draw quickly, but not necessarily accurately.
Archify will check the data structure, node layout, labels, and connection conflicts before delivery.
If verification fails, it identifies the specific issues and suggests allowable fixes, rather than delivering a visually appealing but actually flawed image.
In addition, it can compare two architecture snapshots to highlight additions, deletions, modifications, moves, and reroutes in a Before/Delta/After format.
This means the architecture diagram is no longer just an illustration added after development is complete—it can also be involved in code reviews and evolve alongside the project.
Of course, Archify is not perfect—it is not a real-time monitoring system, nor can it automatically verify that the diagram perfectly matches the online runtime state.
The accuracy of the result still depends on the code information, user prompts, and the Agent's analytical capabilities.
After hands-on testing, some netizens also pointed out that Archify has minor issues, such as the displayed node details being too small, slightly sluggish performance when using models without built-in visual capabilities (e.g., deepseek-v4-flash), and relatively high credit consumption.
In response, the developers, also known as "Wu Feng Yu Wu Qing," said they are working hard to update the version and resolve these issues—stay tuned for an upcoming release.
So, Archify is perfect for you if:
Already using AI for programming, aiming to quickly understand unfamiliar repositories, map out call chains, or explain complex systems to your team.
But Archify's story is not just about how a tool rapidly gained popularity in four months.
There is also a more specific issue behind it:
When your educational background can't be changed, how else can a developer prove themselves?
The answer given by “Neither wind nor rain, nor fog nor clear sky” is to create a work that anyone can open, use, and verify.
I’m a transfer student, but I can do it!
The teacher, "Also no wind, no rain, no fog, just clear skies," has not publicly disclosed their name or current employer, but now Archify’s story has emerged clearly and gained increasing recognition.
Now that the clouds have cleared and the sky is bright, let’s affectionately call him “Teacher Sunny.” (We’ve already made contact—folks interested can stay tuned for the upcoming interview.)
Teacher Qingtian sincerely connected with everyone on Twitter and openly shared his journey—from graduating with a diploma in front-end development, facing numerous setbacks, to ultimately landing an offer from a major Shanghai tech company through his open-source projects and tools. Key themes include:
Started from the countryside, upgraded from associate to bachelor’s degree, self-taught front-end development, repeatedly faced obstacles during background checks, proved my skills through open source contributions.
After the 2018 college entrance exam, Teacher Qingtian enrolled in a vocational program to study computer application technology, and that year he felt his condition was at its worst...
Later, he entered Chongqing University of Posts and Telecommunications' Software Engineering program with the top score in his admission batch.
During the internship, Teacher Qingtian was very happy, and he felt:
As long as I keep working hard, the future will definitely get better.
But the autumn recruitment season dealt him a harsh reality check:
During an internship interview at Westlake University, when the HR brought up his associate-to-bachelor degree again, he finally snapped:
Coming from humble beginnings is not shameful!
In the end, he passed.
But this was only an internship; throughout Qingtian’s job search, no company was willing to offer him a full-time position.
From last year through March of this year, he had to continuously move between internships to gain experience:
Later, he finally met the first mentor of his career—Cz Chen, then Chief Operating Officer of Manus.
△
She suggested that he might demonstrate his abilities elsewhere through his work or open-source projects.
This statement not only confirms the idea that Sunny has long held within their heart:
Your educational background can't be changed anymore; no matter how much you explain, people will still see it as a junior college to bachelor’s upgrade.
Also became a foreshadowing element in Archify’s later story.
So he embarked on an open-source journey, turning his projects into his second resume along the way:
In the end, he landed a job offer from a major company thanks to the foundation he had built himself:
Later, Archify became popular, and he finally seized this opportunity to share with everyone the deep-seated resentment and confusion he had long carried within him:
Open source hasn't turned me from a junior college graduate into a 985 graduate, nor can it guarantee that I won't face barriers due to my educational background in the future—I'm still unsure what these stars will ultimately bring me.
But in the end, he sighed:
At least for now, people are actually using Archify; the fact that what I’ve built can bring value to others and help many people is, to me, the greatest meaning of open source.
Beyond his formal education, he has built a new resume through his own efforts—one written by his actual work, users, and community feedback.
His story has brought hope to more students who have faced repeated setbacks due to insufficient educational qualifications.
And the story of Archify continues.
This article is from the WeChat public account "Quantum Bit," authored by: Focused on Frontier Technologies