| name | omtx - Matrix Generation |
| description | Generates comparison and analysis matrices. Takes items and dimensions, fills cells with evidence-based assessments, and highlights patterns across the matrix. |
| output | {"format":"prose"} |
Matrix Generation
Input: $ARGUMENTS
Step 1: Identify Rows
Determine what is being compared. Rows are the items, options, or entities under analysis.
ROWS (items being compared):
1. [item 1]
2. [item 2]
3. [item 3]
...
Rules:
- Items should be at the same level of abstraction
- 3-7 rows is ideal; more than 10 becomes unwieldy
- If too many items, group or filter first
Step 2: Identify Columns
Determine the dimensions or criteria for comparison.
COLUMNS (dimensions/criteria):
1. [criterion 1] — [what it measures]
2. [criterion 2] — [what it measures]
3. [criterion 3] — [what it measures]
...
Rules:
- Columns should be independent (not redundant)
- Each column must be assessable for every row
- Include both quantitative and qualitative dimensions where relevant
- Weight columns if some criteria matter more than others
WEIGHTS (if applicable):
- [criterion]: [weight or priority level]
Step 3: Fill Cells
Populate each cell with an evidence-based assessment.
Rules:
- Use consistent scales (High/Medium/Low, 1-5, Yes/No, or specific values)
- State the basis for each assessment — no unsupported ratings
- Flag cells where data is uncertain or missing
- Avoid false precision: "~70%" is better than "72.3%" without data
SCALE: [the rating system used]
| | [Col 1] | [Col 2] | [Col 3] | ... |
|---|---|---|---|---|
| [Row 1] | [assessment] | [assessment] | [assessment] | |
| [Row 2] | [assessment] | [assessment] | [assessment] | |
| [Row 3] | [assessment] | [assessment] | [assessment] | |
Step 4: Highlight Patterns
Analyze the completed matrix for structural insights:
| Pattern | What to look for |
|---|
| Dominance | One row scores best on most/all columns |
| Trade-offs | Rows that win on some columns but lose on others |
| Clusters | Groups of rows with similar profiles |
| Outliers | A row that scores very differently on one column |
| Deal-breakers | A single cell that eliminates a row regardless of other scores |
| Parity | Rows that are effectively identical across all columns |
PATTERNS FOUND:
- [pattern 1]: [description]
- [pattern 2]: [description]
...
Step 5: Summarize Findings
Synthesize what the matrix reveals:
MATRIX SUMMARY:
STRONGEST OPTION: [row] — [why]
KEY TRADE-OFF: [row A] vs [row B] — [what you give up for what you gain]
SURPRISING FINDING: [anything the matrix revealed that wasn't obvious]
DECISION IMPLICATION: [what this matrix suggests you should do]
If no single option dominates, state that clearly and explain what additional information or criteria would break the tie.
Integration
Use with:
/olst -> Generate the list of items before building the matrix/cmp -> Deep-dive comparison on the top 2-3 options/odec -> Turn matrix findings into a decision recommendation/cba -> Add cost-benefit analysis to the matrix dimensions
