business-analysis
Business analysis and change router: process modeling, requirements, feasibility, scenarios, enterprise change management, and workshops. Topics: business-process-modeling, change-management-enterprise, change-readiness, feasibility-validation, flow-mapping, requirements-engineering, scenario-analysis, workshop-design, workshop-facilitator.
business-analysis
Router skill. Resolve topic, then Read EXACTLY ONE playbook from routes:. [DOC]
When to use
Analyzing the business side of a system: how work flows, what it must do, whether a path is viable, which option wins, or how people adopt the change. Pick by intent: [INFERENCE]
| Signal in request | topic |
|---|---|
| "how does process X run", BPMN, swimlanes | business-process-modeling |
| "end-to-end flows", DDD contexts, integration map | flow-mapping |
| "what must it do", user stories, acceptance criteria | requirements-engineering |
| "can we / should we", build-vs-buy, constraints | feasibility-validation |
| "compare options A/B/C", weighted trade-off | scenario-analysis |
| "design the session", event storming, story mapping | workshop-design / workshop-facilitator |
| org rollout, stakeholder buy-in, adoption risk | change-management-enterprise / change-readiness |
Routing contract
- Infer
topicfrom the request; ask ONLY if two topics tie. [DOC] - Read the single mapped playbook. NEVER load the whole cluster — defeats the router. [INFERENCE]
depth=deep→ apply the playbook exhaustively, verify at each step;quick→ essentials only. [DOC]- Stack is Firebase/Google-constrained for feasibility/scenario topics — honor it, don't propose off-stack. [CONFIG]
Spine: Discover → Analyze → Execute → Validate.
Quality gates: constitution v6.0.0 (enforcement), evidence tags (Alfa core set,
EN spelling: [DOC]/[INFERENCE]/[ASSUMPTION]/[CONFIG]/[CODE]), script-first rule. [CONFIG]
Acceptance (gate before "done")
- Run the gate in
assets/checklist.md; score againstassets/quality-rubric.json. [DOC] - Exactly one playbook loaded; topic matches user intent. [DOC]
- Every non-obvious claim tagged, one family, consistent spelling. [DOC]
- Outputs (flows, matrices, requirements) trace to evidence, not invention. [ASSUMPTION]
Anti-patterns
- Loading multiple playbooks "to be safe" — pick one. [INFERENCE]
- Guessing
topicwhen genuinely ambiguous instead of asking. [ASSUMPTION] - Pricing or off-stack tooling in feasibility/scenario output. [CONFIG]
- Repository
JaviMontano/jm-adk-beta
- Revision
fdad39c
- Last updated
- First committed
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.