Capstone Evaluation Criteria

A rubric turns the general request to “show the project” into transparent dimensions, evidence expectations, and a decision boundary. A criterion should describe observable quality, not an impression based on code volume or polish.

Evaluation must be separated from unverified claims. Having a rubric does not mean that an evaluator applied it, and a completed example containing Unknown is not a score or approval.

Example dimensions

Dimension	Evidence question	Boundary
Problem and user	Is it clear for whom and which problem the slice solves?	Does not prove market demand or adoption
Core flow	Can the scenario be followed from entry to output?	Does not prove production scale
Scope discipline	Are non-goals and deferred decisions visible?	Do not reward unlimited feature scope
Verification	Are checks connected to acceptance?	A planned check is not a pass result
Evidence and limits	Can the reviewer find facts and Unknowns?	Do not turn Unknown into a score without a decision
Communication	Does the defense have a coherent narrative?	Polish does not replace working evidence

Rubric record

CAPSTONE_RUBRIC.md — documented example

# CAPSTONE_RUBRIC.md
Dimension: <problem / flow / scope / verification / evidence / communication>
Expected signal: <observable criterion>
Evidence required: <file, command or demo anchor>
Status scale: <not shown / partial / shown / needs verification>
Weight: <agreed value or Unknown>
Reviewer note: <feedback or Unknown>
Decision boundary: <GO / HOLD / remediation>
Status: documented example; rubric application and score: Unknown.