Skip to content

ROADMAP.md

Latest commit

 

History

History
37 lines (25 loc) · 2.28 KB

ROADMAP.md

File metadata and controls

37 lines (25 loc) · 2.28 KB
Raw

Roadmap

What is planned, what comes next, and what might happen eventually. For what already shipped, see CHANGELOG.md.

Now

What is done or actively being worked on for the next release.

  • Text pipelines as a dedicated project type / track
  • Building AI into your app: how to integrate LLMs (Claude, OpenAI, local models) into the software you are building - API basics, prompt patterns for in-app use, cost and safety
  • Information flow guide: how data moves through a program (inputs, state, events, outputs), information architecture for AI-assisted projects

Next

Planned, but not started yet.

  • Frontend/UI guide: how React/Vue/Svelte components map to the domain/adapters model
  • Language mapping tables (JavaScript, TypeScript)
  • eslint-plugin-boundaries config (JavaScript/TypeScript)
  • CONTRIBUTING.md: contribution workflow when PRs open
  • Language selection guide for vibecoders: which programming language fits which kind of project (web app, script, data tool, mobile, game). Focused on vibecoding trade-offs: AI support quality per language, ecosystem maturity, deployment simplicity, not on benchmarks or theory.

Later

Ideas we want to explore. No timeline, no commitment.

  • Community-contributed language packs
  • Video walkthrough of the three stages
  • API design guide: how to structure endpoints following the domain/adapters model
  • Database guide: how to choose a database (SQLite vs Postgres vs Redis and when) and how to evolve the schema over time (migrations) without losing data
  • More languages based on demand (C++, Go, Lua)
  • Project templates (cookiecutter/copier per language/stage), likely unnecessary as AI scaffolding improves
  • End-user documentation guide: user guides, feature descriptions, getting-started pages, and in-app help (tooltips, contextual help, onboarding flows). Explicitly out of scope in auto-documentation.md, needs its own guide with different failure modes.
  • RAG system guide: how to structure a retrieval-augmented generation setup (vector store, embeddings, retrieval, prompt assembly) mapped to the domain/adapters model.
  • Working with autonomous coding agents: how you and a long-running agent share one codebase - hand-off points, review gates, what the agent may do on its own, keeping AGENTS.md as the shared contract.