rdra-ish-buc-analyze

name	rdra-ish-buc-analyze
description	Choose the right RDRA-ish BUC or whole-model analysis path for refinement readiness, coverage gaps, business flow, requirement/NFR traceability, screen field mappings, API contracts, access, system boundaries, exports, and state patterns
license	MIT

Choose BUC Analysis

Choose the analysis path that matches the user's question and the model's current refinement stage. Keep this top-level skill as the routing layer, then load the reference file for the concrete analysis task.

Routing Guide

User intent	Analyze first	Load
"Where are we in refinement?"	Stage signals and next missing information	references/stage-readiness.md
"What is incomplete in this BUC/model?"	BUC, actor, use-case, flow, field, CRUD, inferred actor inputs, requirement/NFR scope, and layout gaps	references/coverage-gaps.md
"Are permissions/media correct?"	screen/access paths, permission callables, actor grants	references/access-permissions.md
"Are API/system boundaries sound?"	API ownership, API CRUD, cross-system relations, coordination	references/api-system-boundaries.md
"Are lifecycle states/rules valid?"	reachable state patterns, terminals, truncation, per-entity rule diagnostics, and cross-entity rule inventory	references/state-patterns.md
"Do Int/now/temporal/multi-entity rules hold?"	Prefer skill rdra-ish-verify (TLA+/TLC); else TLA export notes here	references/state-patterns.md
"Summarize findings for review"	severity, evidence, next questions, and concrete recommendations	references/reporting.md

Default Workflow

1. Run rdra-ish check <src-dir>/ first unless the user only wants a narrow command explanation.
2. Identify whether the model is still business-facing or has entered technical commitment: data touchpoints, interaction boundary, entity structure, lifecycle, or rules.
3. Load one reference file from the routing guide and run only the commands needed for that analysis.
4. Report evidence with command output signals, not just intuition.
5. End with the next useful refinement question or the smallest concrete model change.

Quick Command Palette

rdra-ish check src/
rdra-ish lint src/ --format table
rdra-ish list src/ --kind buc --format table
rdra-ish list src/ --kind usecase --format table
rdra-ish list src/ --kind actor --format table
rdra-ish list src/ --kind requirement --format table
rdra-ish list src/ --kind nfr --format table
rdra-ish list src/ --kind field --format table
rdra-ish list src/ --kind api --format table
rdra-ish list src/ --kind adr --format table
rdra-ish csv src/ --kind matrix
rdra-ish csv src/ --kind business-inputs
rdra-ish csv src/ --kind api-matrix
rdra-ish diagram src/ --kind business-area --format mermaid
rdra-ish diagram src/ --kind technical-area --format mermaid
rdra-ish csv src/ --kind screen-constraints
rdra-ish csv src/ --kind permission-callables
rdra-ish csv src/ --kind actor-permission-audit
rdra-ish states src/
rdra-ish states src/ --buc <BucId>
rdra-ish states src/ --entity <EntityId>
rdra-ish export src/ --kind tla -o out/
rdra-ish export src/ --kind openapi --out out/openapi.json
rdra-ish export src/ --kind dbml --out out/schema.dbml

Reference Files

• references/stage-readiness.md — classify the current abstraction stage and decide what information should be requested next.
• references/coverage-gaps.md — find missing BUC ownership, actor/use-case links, CRUD touchpoints, layout issues, and import hygiene.
• references/access-permissions.md — analyze permission/media requirements and actor grant coverage with the three access CSVs.
• references/api-system-boundaries.md — review API invocation, system ownership, API/entity matrices, and cross-system coordination.
• references/state-patterns.md — interpret states, lifecycle warnings, terminal/unreachable patterns, truncation, and rule diagnostics. For full TLA+/TLC workflow, load skill rdra-ish-verify instead.
• references/reporting.md — structure analysis output with severity, evidence, and recommendations.