spd-space-discovery

name	spd - Space Discovery
description	Discover what space exists BEFORE generating guesses. Maps dimensions, stakeholders, time horizons, and adjacent spaces. Run this before guess_generation to ensure comprehensive coverage.
output	{"format":"prose"}

Space Discovery

Input: $ARGUMENTS

---

Core Principle: DISCOVER BEFORE COVERING

Don't assume the space is known. Discover it first. Guesses CREATE the space - they don't map to a pre-existing space.

This procedure runs BEFORE /gg to define what needs to be covered.

---

Step 1: What COULD This Be About?

List all possible framings of the input:

Frame	How It Changes the Space
Literal interpretation	[what space does this create?]
Metaphorical interpretation	[what space does this create?]
Surface-level reading	[what space does this create?]
Deep reading	[what space does this create?]
Best-case framing	[what space does this create?]
Worst-case framing	[what space does this create?]
User's likely intent	[what space does this create?]
Alternative intents	[what space does this create?]

Frames discovered: [list all, aim for 5-10]

---

Step 2: Who COULD Care About This?

List all stakeholders who might be affected or have perspectives:

Stakeholder	Their Perspective	Why It Matters
The speaker/user	[their view]	[impact on space]
Direct beneficiaries	[their view]	[impact on space]
Direct losers	[their view]	[impact on space]
Implementers	[their view]	[impact on space]
Future selves	[their view]	[impact on space]
Society/public	[their view]	[impact on space]
Adversaries	[their view]	[impact on space]
Neutral observers	[their view]	[impact on space]

Stakeholders discovered: [list all, aim for 5-10]

---

Step 3: What Time Frames COULD Matter?

List all relevant time horizons:

Time Horizon	What Changes	Relevance
Immediate (now)	[what's true now?]	[why it matters]
Short-term (days/weeks)	[what changes?]	[why it matters]
Medium-term (months)	[what changes?]	[why it matters]
Long-term (years)	[what changes?]	[why it matters]
Historical context	[what led here?]	[why it matters]
Cyclical patterns	[what repeats?]	[why it matters]
Irreversible changes	[point of no return?]	[why it matters]

Time horizons discovered: [list all relevant ones]

---

Step 4: What Dimensions COULD Vary?

For the input, identify ALL axes of variation:

Standard Dimensions (check each)

Dimension	Applicable?	Possible Values
AGENT (who)	YES/NO	[enumerate if yes]
ACTION (what doing)	YES/NO	[enumerate if yes]
OBJECT (what affected)	YES/NO	[enumerate if yes]
REASON (why)	YES/NO	[enumerate if yes]
METHOD (how)	YES/NO	[enumerate if yes]
TIME (when)	YES/NO	[enumerate if yes]
LOCATION (where)	YES/NO	[enumerate if yes]
DEGREE (how much)	YES/NO	[enumerate if yes]
CERTAINTY (how sure)	YES/NO	[enumerate if yes]
SCOPE (how broadly)	YES/NO	[enumerate if yes]

Domain-Specific Dimensions

For this specific input, what OTHER dimensions exist?

Custom Dimension	Possible Values
[dimension 1]	[values]
[dimension 2]	[values]
[dimension 3]	[values]

Dimensions discovered: [count of applicable standard + custom]

---

Step 5: What Adjacent Spaces COULD Connect?

What related problems/topics connect to this one?

Adjacent Space	Connection	Why It Matters
[related topic 1]	[how it connects]	[might need to explore]
[related topic 2]	[how it connects]	[might need to explore]
[related topic 3]	[how it connects]	[might need to explore]
[prerequisite]	[must be true for this to matter]	[check first]
[consequence]	[follows from this]	[downstream effect]

Adjacent spaces discovered: [list]

---

Step 6: What Types of Claims COULD Exist?

Check which claim types are relevant:

Claim Type	Applicable?	Example in This Context
Factual	YES/NO	[example if yes]
Causal	YES/NO	[example if yes]
Predictive	YES/NO	[example if yes]
Normative	YES/NO	[example if yes]
Modal	YES/NO	[example if yes]
Relational	YES/NO	[example if yes]
Intentional	YES/NO	[example if yes]
Meta	YES/NO	[example if yes]

Claim types applicable: [list]

---

Step 7: Synthesize the Space Map

Combine all discoveries into a space map:

SPACE MAP FOR: "[input]"

FRAMES: [count] possible interpretations
├── [frame 1]
├── [frame 2]
└── ...

STAKEHOLDERS: [count] perspectives to consider
├── [stakeholder 1]
├── [stakeholder 2]
└── ...

TIME HORIZONS: [count] relevant
├── [horizon 1]
├── [horizon 2]
└── ...

DIMENSIONS: [count] axes of variation
├── Standard: [which apply]
├── Custom: [domain-specific]
└── Total cells: [product of values]

ADJACENT SPACES: [count] connections
├── [adjacent 1]
├── [adjacent 2]
└── ...

CLAIM TYPES: [count] / 8 applicable
└── [list applicable]

TOTAL SPACE SIZE ESTIMATE:
[frames] × [stakeholders] × [time horizons] × [dimension combinations] = [rough size]

---

Step 8: Recommend Coverage Strategy

Based on space size, recommend coverage mode for /gg:

Space Size	Recommended Mode	Rationale
Small (<100 regions)	EXHAUSTIVE	Can cover everything
Medium (100-1000)	ADAPTIVE	Start broad, go deep on signal
Large (1000-10000)	TARGETED	Focus on high-impact regions
Huge (>10000)	BOUNDARY	Map edges, sample interior

Recommended mode: [MODE] Reason: [why this mode]

---

Output Format

## SPACE DISCOVERED FOR: "[input]"

### Frames: [N]
[list]

### Stakeholders: [N]
[list]

### Time Horizons: [N]
[list]

### Dimensions: [N standard + M custom]
Standard applicable: [list]
Custom discovered: [list]

### Adjacent Spaces: [N]
[list]

### Claim Types: [N/8]
[list applicable]

### Space Size Estimate
[calculation]

### Coverage Recommendation
Mode: [EXHAUSTIVE/ADAPTIVE/TARGETED/BOUNDARY]
Reason: [why]

### Priority Regions
High-impact regions to cover first:
1. [region]
2. [region]
3. [region]

### Skip Candidates
Regions that might be skipped:
1. [region] - [reason to skip]
2. [region] - [reason to skip]

---

Next Procedure

→ INVOKE: /gg [input] --coverage-mode=[MODE] --space-map=[this output]

---

Execute now: Discover the space before guessing.