lamina-evaluation

Judge product usability and evidence quality. Use when running Persona-based walkthroughs on a built product, planning an expert heuristic review, defining evidence-backed success metrics, or checking

SKILL.md

GitHub

Lamina Evaluation

Reference-loading protocol

  1. Match the request's primary evaluation method to one row below.
  2. Open that linked reference before answering. Add another only when a second method materially changes the answer; do not preload the directory.
  3. Start the response with Using lamina-evaluation: <topic path(s)> so the selected evaluation lens is auditable.

Topic index

Evaluation signal Read Adds
Need actors to attempt workflows and edge probes on a built product Actor Evaluation Persona-based walkthrough method and reproducible blockers
Need specialists to inspect a contract or live product through explicit lenses Expert Lens Review parallel lens coverage tied to evidence and contract refs
Need success metrics, analytics interpretation, or experiment claims Metrics Discipline measurement boundaries and anti-fabrication rules

Working rule

Use the smallest sufficient reference set. Actor evaluation requires a runnable product; use expert review before build or when a lens-specific inspection is requested. Pair either with metrics discipline only when real measurements or a measurement plan are in scope.