Tried tt-a1i/archify: A Practical Diagram Agent for AI Teams

I took tt-a1i/archify for a quick test drive today. It is an agent skill for generating beautiful, verifiable architecture, workflow, sequence, data-flow, and lifecycle diagrams—not just static boxes and arrows.

The output is self-contained HTML with motion, readable layouts, and crisp export options. That matters in real reviews: diagrams can live in a pull request, be opened offline, and remain useful during an incident or design walkthrough. The project is also gaining serious traction, with +4,562 GitHub stars today.

For an enterprise setup, I would route the model traffic through a controlled OpenAI-compatible gateway rather than handing provider credentials directly to every developer. For example:

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["B_LOST_API_KEY"],
    base_url="https://b-lost.com/v1",
)

response = client.chat.completions.create(
    model="claude-fable-5",
    messages=[
        {
            "role": "user",
            "content": (
                "Create a verifiable sequence diagram for an order workflow. "
                "Return self-contained HTML with animated transitions."
            ),
        }
    ],
)

print(response.choices[0].message.content)

This pattern gives platform teams one place to enforce team token quotas, routing rules, audit boundaries, and private-network egress policies. It also keeps the integration compatible with tools such as Cursor, Cline, Roo Code, Windsurf, Aider, LibreChat, and other OpenAI-style clients.

The B-Lost Universal Relay uses https://b-lost.com/v1, advertises 20% off official list pricing, and supports native Anthropic /v1/messages prompt caching. For repeated architecture context—standards, naming conventions, service catalogs—full prompt caching with a 90% cache-hit discount can reduce both latency and spend. I would still validate the model alias and retention policy before production deployment, especially when diagrams contain internal topology or customer data.

archify is worth watching if your team wants documentation that is both presentation-ready and technically inspectable.