v0.1.0-alpha.4

Fourth pre-release of ycore-ddd. A new project is set up in one call, the module names its starting point, and bmad-ux can be bound to the domain model.

What changed

• Installation (#22). The quick start installs BMad with its core-tools and method skills, so bmad-spec, bmad-build, bmad-ux and bmad-retrospective are there after the domain model.
• Quick start in one call (#22, #26). One claude call sets up git, runs bmad setup and makes both bindings.
• Starting point (#23, #24). The post-install message names /ycore-ddd-big-picture as the starting point. With only an idea description, the bmad agent now offers both routes with their reasons: straight to the Big Picture when the open questions are about structure, or a product brief first when the business side is open.
• UX bound to the domain model (#26). /ycore-ddd-big-picture bind the ux to the domain model has bmad-customize write a team override for bmad-ux: every UX run loads glossary, Context Map, event flow and designs, uses the glossary's terms strictly, and lays out screens and flows along the Bounded Contexts. The README shows bmad-ux in the BMad flow.

Install or update

Requires BMad 6.13 (6.13.0-next from the BMAD-METHOD repository); the full install command is in the README.

npx skills update -p      # update an existing project, then start a new session

In a project set up before this release, make the UX binding once: claude "/ycore-ddd-big-picture bind the ux to the domain model".

Tested so far

Run successfully in real runs: Big Picture, Design Level and the model check on the working example, the Fast path with one final acceptance, the retrospective binding in a fresh project, continuing an interrupted run, and "Set up git for this project and run bmad setup" in an empty folder.

Not yet exercised: the stakes question and Hotspot sorting, the UX binding with a following bmad-ux run, and the combined quick-start call that also makes both bindings.

v0.1.0-alpha.3

Third pre-release of ycore-ddd. The skills now start reliably when asked in your own words, and the README shows what a run looks like.

What changed

• Skills: trigger evals and scoped descriptions (#19). Each skill has evals/triggers.json with ten queries that should start it and ten near misses, German and English. Each description now also says what the skill is not for. Measured with bmad-eval (3 runs per query): 49 of 60 queries pass, up from 42; every query that should start a skill does so in every run. The remaining misses are near misses that a neighbouring skill would take in a real project; the eval stages the skill under test alone.
• README: a run and a quick start (#20). A shortened Big Picture run in the Fast path and a table of phrasings that start each skill now open the README.
• README: the trial run is complete (#18). The first trial run of the working example reached the model check and the bound retrospective.

Install or update

Requires BMad 6.13 (6.13.0-next from the BMAD-METHOD repository), see the README.

npx skills add y-core-engineering/bmad-ddd-module -a claude-code   # first install
npx skills update -p                                               # update

Start a new session after updating so the new descriptions are read.

Tested so far

Run successfully in real runs: Big Picture, Design Level and the model check on the working example, the Fast path with one final acceptance, the retrospective binding, and continuing an interrupted run in a new session. Not yet exercised: the stakes question and Hotspot sorting.

v0.1.0-alpha.2

Second pre-release of ycore-ddd. The skills are unchanged since v0.1.0-alpha.1; this release updates the documentation.

What changed

• README: badges for version, tests, licence, BMad 6.13, the tested agent and uv; the requirements stated in one place (#15)
• README: the diagrams are rendered to SVG from Mermaid sources in docs/diagrams/ (#16)

Install or update

Requires BMad 6.13 (6.13.0-next from the BMAD-METHOD repository), see the README.

npx skills add y-core-engineering/bmad-ddd-module -a claude-code   # first install
npx skills update -p                                               # update

Since alpha.1 the Fast path with one final acceptance and the retrospective binding have run successfully in real runs. Not yet exercised: the stakes question and Hotspot sorting; trigger evals are being added.

v0.1.0-alpha.1

First pre-release of ycore-ddd, the domain modelling part of the Y-Core Method as a BMad community module.

Requires BMad 6.13

The module is built and tested against BMad 6.13.0-next from the BMAD-METHOD repository. 6.13 is not yet released on npm (latest there: 6.12.2), so install BMad from the repository, not with the bmad-method installer:

npx skills add bmad-code-org/BMAD-METHOD --skill bmad --skill bmod-core-tools -a claude-code
npx skills add y-core-engineering/bmad-ddd-module -a claude-code
claude "bmad setup"

Projects set up with BMad 6.12 or earlier need bmad migrate method first. Should BMad change formats before 6.13 is released, this alpha may need an update.

What is in it

• ycore-ddd-big-picture: prepares and facilitates a Big-Picture Event Storming with a team of agents; writes context-map.md, ubiquitous-language.md and event-storming.md.
• ycore-ddd-design-level: builds a catalog of the Bounded Contexts and runs a Design-Level Event Storming per context into design.md (Commands with permissions, Aggregates with invariants, abuse cases).
• ycore-ddd-model-check: checks the code against the model in a fresh-context subagent, then decides every high and medium finding with you; can bind itself to bmad-retrospective.

The workshops know create, update and validate, a Coaching and a Fast path, a reviewer gate, headless runs, and they put only domain questions to you. Their output feeds bmad-spec, and bmad-prd / bmad-architecture where the work needs them. See the README for usage and a working example.

Tested so far

One end-to-end trial run on the hello world example: Big Picture done, Design Level done (one lean context), model check final with all decisions. All artifacts pass the module's check scripts.

Not yet exercised in a real run: the stakes question and Hotspot sorting, the Fast path with one final acceptance, and the retrospective binding. Trigger evals have not been run yet.