Product Management composition examples

See bounded Product Management journeys that hand off to existing Adaptive Skills contracts.

These examples show sequencing, not automatic orchestration. Each handoff requires the receiving skill’s inputs and preserves human authority over the decision.

1. Requested feature with no problem

1. product-management activates problem-framing.
2. The output separates the requested solution from the user problem, desired outcome, context, and evidence gaps.
3. If the value mechanism is still unclear, use revenue-lever-mapping or feature-value-governance after the frame is accepted.
4. Only a bounded change moves to feature-planning.

The safe result is a framed question and next evidence step, not a feature approval or delivery date.

2. Data-poor prioritization

1. product-management classifies the decision as comparative sequencing with incomplete evidence.
2. prioritization-and-method-selection rejects unsupported RICE or ICE precision.
3. The team uses an explicit comparison with visible assumptions, dependencies, and human-owned weights.
4. feature-value-governance or feature-complexity-audit can deepen a specific investment question.

The output is a decision-ready comparison with uncertainty, not an invented ranking.

3. Activity metric presented as value

1. metrics-and-outcomes labels the reported activity signal and separates it from the intended outcome.
2. The team defines a measurement window, outcome signal, and guardrail.
3. observability-review checks whether the required events and diagnostics can support the claim.
4. A human owner decides whether the evidence supports continuation, change, or stop.

A composition rule

When a request contains multiple questions, decompose it before loading modules. Each handoff should state its input, owner, expected result, and verification. If the next step crosses investment, roadmap, or macro-governance authority, escalate to the consumer governance layer or AletheIA instead of extending the skill’s authority.