30
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill lyndonkl/claude --skill decision-matrix- SKILL.md8.9 KB
Overview
This skill helps teams compare multiple named alternatives across weighted criteria to produce a transparent, data-driven recommendation. It makes trade-offs explicit by scoring options against agreed criteria, calculating weighted totals, and highlighting sensitivity and assumptions. Use it when decisions must be defensible, documented, and easy to share with stakeholders.
How this skill works
You frame the decision, list named alternatives, and capture any must-haves or constraints. Next you define criteria and assign weights (percentages summing to 100%), then score each option on a consistent scale (typically 1–10). The skill calculates weighted scores, ranks alternatives, runs basic sensitivity checks (close calls, dominant criteria, weight/score sensitivity) and produces a concise recommendation with rationale and next steps.
When to use it
- Comparing vendors, tools, or suppliers where multiple factors matter
- Choosing between project or product strategies with competing priorities (cost vs quality vs speed)
- Making group decisions that require stakeholder alignment and documented rationale
- When stakeholders ask “which option should we choose” or “compare alternatives”
- When a decision must be defensible, repeatable, and transparent
Best practices
- Frame the decision clearly: name the choice, deadline, and consequences of a wrong choice
- List only specific, named alternatives; filter out options failing must-have constraints before scoring
- Keep criteria to a manageable number (5–8) and ensure weights sum to 100%
- Document data sources and assumptions for every score; flag guesses for follow-up research
- Run sensitivity checks: flag options within ~5–10% of the winner for more validation
- Use stakeholder averaging or pairwise comparison for group weight-setting to surface priority differences
Example use cases
- Vendor selection: price, integration, support, reputation, contract terms with weights based on procurement priorities
- Technology choice: performance, scalability, maturity, developer productivity for platform decisions
- Go/no-go strategic choices: market opportunity, resource needs, risk, alignment with company strategy
- Feature prioritization: user impact, implementation effort, strategic value, dependencies
- Hiring shortlist: experience, culture fit, growth potential, availability
FAQ
Use a consistent numeric scale such as 1–10 where 10 is best; document what each score means to keep scoring consistent across reviewers.
What if the winner is a close call?
If options are within ~5–10% of each other, gather more data on the sensitive criteria, run stakeholder reviews, or perform targeted sensitivity analysis before finalizing.