[About]
About stacktxt: we make websites readable to models.
The web was built for human eyes: layouts, animation, and client-side scripts. AI assistants and shopping agents work differently. They fetch files and read text, so a store can look perfect to a person and still be hard for a model to read.
stacktxt builds the layer they read: crawler rules, plain-text manifests, structured data, and product feeds.
[How we work]
Four operating principles
- 1
Plain text is the protocol
When tokens cost compute, brevity and structure win.
- 2
Work over theory
We ship code, robots rules, and edge workers, not slide decks.
- 3
Calm, verifiable engineering
AI is a reader to format for. We report what we measured.
- 4
No filler
No generated padding, no buzzwords, no fake empathy.
[Founders]
Max and Kellen
Both founders join every review. We are based in Dallas, TX and work with stores across the United States.
Maxwell Thomason
[Co-founder]
Architecture, AI ingestion systems, and technical data pipelines.
Kellen Johnson
[Co-founder]
Commercial strategy, growth operations, and client deliverables.
[Scope]
What we do not do
A narrow scope keeps the work verifiable.
- Storefront redesigns We build the machine-readable layer, not human-facing themes.
- Blog and marketing copy We write dense, structured files for models to read.
- Ad account management We do not run ad spend, creative, or bidding.
- ERP and warehouse software We integrate with standard platforms and do not rebuild internal systems.