Score a spec/requirement on 4 axes. YOU (the LLM) read the spec and score it.

Score the spec on these 4 axes (each 0.0-1.0):

• completeness: Can an LLM understand the full scope? (0=vague idea, 1=complete with goal+context+criteria)

• clarity: Is it unambiguous? (0=multiple interpretations, 1=one clear interpretation)

• constraints: Are boundaries defined? (0=no limits, 1=clear scope and non-goals)

• specificity: Concrete testable details? (0=no testable criteria, 1=verifiable outcomes)

Also identify the weakest axis and give one tip to improve it.

The tool normalizes scores, calculates balance (0-1), and returns a verdict. A balanced spec produces reliable LLM output. An unbalanced spec produces hallucinations.

Score a spec and generate an SVG radar chart. YOU (the LLM) read the spec and score it on 4 axes (0.0-1.0 each): completeness, clarity, constraints, specificity. Also identify weakest axis and tip.

The tool generates a radar chart: green = balanced (ready), yellow = moderate (gaps), red = spiked (blind spots).

Side-by-side radar charts comparing two specs. YOU (the LLM) score BOTH specs on 4 axes each. Useful for before/after or good/bad examples.