Supplier Reply Decoder
Paste any supplier email. The scan runs in your browser — bilingual pattern detection for vague commitments, dodge phrases, pressure tactics, and payment red flags — then builds a custom AI prompt that analyzes the reply line by line.
Nothing leaves your browser. The generated prompt goes to your AI session, not ours.
The four pattern classes
| VAGUE | Commitment-shaped words with no date, number, or owner attached — "no problem", "we will try". |
| DODGE | Deferral chains — "check with our engineer" — every hop adds a day of silence and removes a named accountable person. |
| RISK | Payment-account changes and manufactured urgency. The deposit-scam signature class. |
| SIGNAL | Identity tells — "our factory / our workers" vs "our partner factory" — factory voice vs middleman voice. |
Why words beat feelings in supplier email
Every importer knows the feeling: a warm, friendly supplier reply that somehow commits to nothing. The warmth is real — Chinese business culture is relationship-forward — but warmth and commitment are different currencies. "No problem" (没问题) in a supplier email is not a yes; a yes is a PI with the spec, date, and price in writing.
The patterns this tool detects are the ones that cost importers money when left unread: the vague commitments that evaporate at QC time, the deferral chains that burn weeks, the urgency theater that pressures you past verification, and the account changes that reroute your deposit to a person instead of a company.
The full decoder — 12 high-stakes phrases with the Chinese side of the table explained — is in the phrase decoder guide. The complete system (25 templates, verification protocol, payment-safety tiers) is The China Sourcing Playbook.
FAQ
Does the tool send my email anywhere?
No. Pure client-side JavaScript. The scan, the scoring, and the prompt generation all run in your browser tab.
Why generate a prompt instead of analyzing it here?
Two reasons: your email stays between you and your AI provider (this page never sees it server-side), and the line-by-line judgment call — what's a cultural norm vs what's a dodge — is exactly what LLMs are good at, once the pattern scan tells them where to look.