Assess named artifacts with evidence-cited Hermes Rubric results and explicit coverage limits.
Assess an agent output or application artifact with cited evidence, coverage facts, dimension scores, and caller-controlled feedback using the Hermes Rubric CLI or Python API.
