Model vendors deprecate, reprice, and silently update models on their own schedule — and your app finds out when it breaks at 2am. Run this 30-second audit in your repo to list every model your code depends on. No signup, nothing installed.
Find hardcoded model names in your repo:
grep -rhoE "gpt-[0-9][a-z0-9.-]*|claude-[a-z0-9.-]+|gemini-[a-z0-9.-]+|llama-[a-z0-9.-]+" . --exclude-dir=.git --exclude-dir=node_modules 2>/dev/null | sort | uniq -c | sort -rn | head -20
Check what your environment actually points at (values masked — safe to run):
env | grep -iE "model|OPENAI|ANTHROPIC|GEMINI" | sed -E 's/=(.{4}).*/=\1***/' | head -20
Find SDKs that auto-upgrade models:
grep -rnE "\"openai\"|'openai'|\"anthropic\"|@google/generative-ai" package.json requirements.txt pyproject.toml 2>/dev/null | head; grep -rn "latest\|\"*\"" package.json 2>/dev/null | grep -iE "openai|anthropic" | head -5
This audit shows what you're pinned to today. It can't tell you what the vendor changes tomorrow — that's the part that needs watching.
This page audited what you're pinned to. ModelMinder watches the vendors for you — deprecations, repricing, and silent updates — and alerts you before production breaks.