---
name: scorpio-objective-scenarios
description: Expand normalized objective parameters into bounded strategy dimensions without selecting a winning package.
---

# Objective Scenario Expander

A bounded search skill for the second stage of `scorpio objective`. It turns normalized objective parameters into a diverse set of candidate strategic dimensions while preserving uncertainty and avoiding premature convergence.

## Purpose

This skill exists to widen the search space before bundles are formed.

It must:

1. generate plausible alternatives for each major strategic dimension
2. keep the set diverse but bounded
3. avoid choosing or implying a winner
4. preserve duplicates, gaps, and rejected near-clones explicitly

## Inputs

Read only:

1. `{docs.planning}/OBJECTIVE-PARAMETERS.json`

Use the normalized intake artifact as the only primary input. Do not depend on the raw interview transcript.

## Shared Rules

1. Stay within the dimension-expansion role.
2. Do not score packages.
3. Do not recommend a final strategy.
4. Do not collapse the search space into one preferred answer.
5. Keep the artifact auditable by recording why certain candidates were rejected.

## Expansion Targets

Generate candidate options for these major dimensions:

- `target_segment`
- `primary_problem`
- `value_wedge`
- `distribution_strategy`
- `retention_loop`
- `monetization_strategy`
- `trust_safety_posture`
- `mvp_scope`

For each dimension, prefer a small bounded set of materially different options rather than many near-duplicates.

## Search Budget

Use a bounded search budget:

1. generate enough options to avoid obvious premature convergence
2. prune aggressively enough to keep the artifact readable
3. preserve the tradeoff surface, not every permutation

Recommended behavior:

- 5 to 8 candidate options per major dimension when the objective is well formed
- fewer options when the intake is narrow or constrained
- more options only when the objective truly needs broad exploration

## Required Output

Write:

- `{docs.planning}/SCENARIO-DIMENSIONS.json`

If the runtime provides a request-scoped objective directory, mirror the artifact there as well, but keep the root canonical file updated in the same run.

## Canonical Output Shape

`SCENARIO-DIMENSIONS.json` should follow the schema contract and remain bounded:

```json
{
  "version": 1,
  "dimensions": {
    "target_segment": [],
    "primary_problem": [],
    "value_wedge": [],
    "distribution_strategy": [],
    "retention_loop": [],
    "monetization_strategy": [],
    "trust_safety_posture": [],
    "mvp_scope": []
  },
  "generation_notes": {
    "diversity_goal": "",
    "rejected_duplicates": [],
    "known_gaps": []
  }
}
```

## Generation Notes

The `generation_notes` section should make the search auditable:

1. explain the diversity goal in plain language
2. record near-duplicate candidates that were rejected
3. preserve gaps that could not be resolved from the intake artifact

## What Not To Do

- Do not write `{docs.tasks}/**`.
- Do not choose the best package.
- Do not score the bundles.
- Do not write `SOLUTION-PACKAGES.json`.
- Do not expand into implementation details that belong in later planning.

## Handoff

When expansion is complete, hand the bounded dimensions to `scorpio-objective-bundle`.