CSD Matrix Agent
Builds a CSD matrix — Certainties, Suppositions, and Doubts — to align the team on what it knows versus assumes.
What it does
The CSD Matrix separates what your team actually knows from what it is assuming. It sorts everything on the table into Certainties (evidence-backed), Suppositions (believed but unverified) and Doubts (open questions) — so the assumptions everyone has been treating as fact become visible.
When to use it
Run it when a decision feels confident but you cannot point at the evidence, or when the team keeps circling the same disagreement. It is most useful before committing to a direction, because it shows exactly which beliefs would be cheapest to test first.
Finding the assumption everyone forgot was an assumption
A PM about to commit a quarter to a rebuild
The situation. The team agrees the current onboarding is the problem, but nobody can say which part, and the evidence is six months old.
What they do
- They run the CSD Matrix against the discovery evidence on the canvas.
- It sorts the claims: only two are Certainties, six are Suppositions, and one Doubt turns out to be load-bearing.
- The biggest Supposition — "users abandon at the invite step" — has no evidence behind it at all.
- They spend a week validating that one belief instead of a quarter rebuilding around it.
The result. The rebuild is scoped to what the evidence supports, and the untested belief is either confirmed cheaply or dropped before it costs a quarter.
What it can do
- Assumption surfacing
- Certainty vs supposition separation
- Doubt-to-research mapping
- Validation planning
What you give it
When you run this agent you can tune it with these settings (sensible defaults are provided):
What you get back
- a Certainties / Suppositions / Doubts matrix, separating what you know from what you assume
Every output lands on your canvas as a node you can edit, connect, and trace back to its source.
Examples
- Build a CSD matrix for this project
- What do we know vs assume?
- Surface our riskiest assumptions
Related agents
Competitor Analysis Agent
Structured competitive analysis — competitor profiles, feature comparison matrices, market positioning maps, and strategic recommendations.
Definition Agent
Synthesizes research into clear problem statements and requirements.
Empathy Map Agent
Clusters user research into distinct personas with goals, frustrations, and motivations.
Librarian Agent
Organizes and retrieves design knowledge and best practices.