Arda Platform Documentation

Documentation for the people who build Arda — and the agents collaborating with them. No fluff, no jargon hunting.

Find your starting point Agent Access GitHub

The Arda Platform turns kanban on the shop floor into a closed-loop control system that small and mid-size manufacturers can actually run. Physical cards scan into digital reorders, planning runs in the background, and stockouts stop being a recurring fire drill. Under the hood: an API-first, MCP-native architecture grounded in operations science — Little’s law, bitemporal state, decoupled control loops. On the surface: a shop-floor experience that prefers cards and scans over screens. This site documents how the platform is built, deployed, run, and extended — for the people doing each of those, and the AI assistants helping them.

Find your starting point

Pick the role closest to yours. Every link drops you at a curated entry point.

Product Manager Personas, use cases, and the roadmap of capabilities the platform supports.

Product or DevOps Engineer Architecture, design patterns, runtime, OAM, and the SRE runbooks that keep production honest.

Customer Service or Sales Engineer Workflows users actually run, plus release notes you can quote back to a customer with confidence.

AI or Tooling Builder The /llms.txt index, raw .md endpoints, and how to point your assistant at this site.

Explore by content area

Product Markets, personas, features, offerings — what the platform does for the customer.

Domain Information model, glossary, the language we use to describe the business.

Current System Architecture, runtime, data model, OAM — how it works today.

Process Engineering craft, project lifecycle, SRE practices, deployment conventions.

Technology Per-technology references: Next.js, Ktor, Astro & Starlight, Exposed, CDK, AG Grid, Cedar.

Vision Long-term view of the mature system once foreseeable roadmap projects are complete.

Decisions Durable choices and the rationale behind them.

Roadmap What is shipping, what is next, and where past projects landed.

🤖 Built for AI-augmented teams

Arda’s teams use AI assistants every day — drafting PRs, drafting runbooks, drafting answers to customer questions. This site is designed so those agents read it as easily as the humans do.

/llms.txt — curated index of canonical entry points. Drop it into any tool that supports the llms.txt convention.
/llms-full.txt — the whole site as a single Markdown bundle for RAG, embedding, or @docs attachments.
/llms-small.txt — hierarchy and per-page descriptions, sized for a tight context window.
/<slug>.md — every rendered page is also served as raw Markdown. Just append .md to any page’s URL.

✍️ Contribute

Every page is plain Markdown in a public GitHub repository. The bar is low, the conventions are documented, and the build catches you when you stray.

Authoring Guide Start here — frontmatter, links, diagrams, the queued CI/CD pipeline.

Frontmatter and Build Rules What the build enforces, what reviewers expect, how to read a CI failure.

Templates Runbooks, design docs, decision logs, ADRs, content pages — start from one.

Markdown Gallery Every component and syntax the site supports, in one place.