diagram-data
Draw the entity/relationship view of the data model from the repo's schema adapter, or the data diff between two refs. Use for "what data do we store and how does it relate", before changing the data model, or to show what a branch changes in it.
- License 许可
- MIT
- 最近更新
- 2026-11-04
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- diagram-data
- metadata
- { "layer": "L3" }
- description
- Draw the entity/relationship view of the data model from the repo's schema adapter, or the data diff between two refs. Use for "what data do we store and how does it relate", before changing the data model, or to show what a branch changes in it.
03 SKILL.md 正文
Diagram data
Show the entities and how they relate, from the schema the repo declares. You choose the scope; the adapter supplies every entity, field and relation.
1. Scope
- One entity and its direct relations, a named group (one workspace's entities), or the whole model when it is small (about 15 entities).
- Show the fields that matter to the question: keys, references to other entities, state fields, and the fields being changed. Write
+n morefor the rest. - Map domain words to entity names with
$(ven-profile path)/GLOSSARY.md.
2. Generate
- Run
ven-profile adapter schema.read. It prints{entities:[{name,fields,relations}]}. - Keep the entities in scope plus one hop of relations. Entities outside the scope appear as stubs, name only.
- Draw each relation with the cardinality the output gives. Never invent one: when it is missing, draw
}o--o{, the least committal form, and say so under the diagram.
erDiagram
ACCOUNT ||--o{ WIDGET : owns
WIDGET }o--o{ REPORT : "appears in"
ACCOUNT {
text id PK
text status
}
WIDGET {
text id PK
text account_id FK
}
Badge it GENERATED ven-profile adapter schema.read @ <short sha>, per the badge protocol in diagram-deps.
3. Data diff between two refs
- Run
schema.readat head, then at the base ref as "At another ref" in the badge protocol describes. - Compare by entity name: entities added or removed, fields added, removed or changed (type, nullability, default), relations added or removed.
- Draw only the changed entities plus their direct neighbours as stubs. Mark each change in the attribute comment (
text email "added",int count "type changed: was text"); draw removed entities with aremovedcomment on their key. - Under the diagram, list each change as one line. Flag the ones that need a data migration or break existing readers.
The badge carries both refs: @ <base sha>..<head sha>. Render it per Rendering in the badge protocol of diagram-deps.
Without the adapter
Find where the data model is declared: schema files, migration files, model definitions (the profile's stack and ven-profile detect point at them). Each entity, field group and relation cites its declaring file:line, so the result is HYBRID under the protocol's no-adapter rule. When migrations are the only source, replay them in order and cite the migration that last touched each field.
04 同仓库更多 Skills
yeet
Take every local change green through the repo's own checks, slice it into conventional commits, push, and with --pr open a pull request on the code host. Use only when the user explicitly asks to yeet, or to commit and push everything; never on your own initiative.
write-docs
Write or update documentation from the code: gather facts, pick the shape for the reader, write it plain, and prove every command and path in it. Use when asked to document a repo, module or feature, or to bring stale docs up to date.
why
Find why code is shaped the way it is (design rationale, trade-offs, regressions, odd thresholds) from history, tickets, docs and chat, citing every claim and stating its confidence. Use for 'why does X work this way', 'why did we pick Y', or before changing code whose purpose is unclear.
verify
Prove a change works with the check that matches its type, do the read-back, and log the pass against the exact tree. Use when a change is ready to call done, before review, commit or stopping, and whenever ven-decisions end refuses an unverified tree.
unslop
Cut AI writing tells (filler, model vocabulary, metaphor nouns, hedging, formatting tics) from text while keeping every fact. Use when editing or checking any prose before it ships: docs, PR bodies, reports, skill text.
type-discipline
Apply when designing types or data shapes, reviewing a signature, or handling outside data, in any language. Make wrong states fail early: in the compiler for typed languages, in validated runtime contracts for untyped ones.