Injection Risk Scorer

Paste brand content to get a Dual-Model Content Safety Score — separate Claude suppression risk and GPT appeal readings with transparent signal breakdown and evidence-led rewrites. Heuristic runs instantly in your browser.

Score a passage

0/8000 characters · 0 words · ~0 tokens

Scoring runs instantly in your browser — no API key required. Nothing is uploaded to richvoller.com. Optionally run a live dual-model probe after your result appears.