aitoubiaoling-bid-review

Reviewing a bid submission against its RFP is painstaking: you must verify eligibility responses, pricing, schedule, personnel, technical parameters, implementation plans, service commitments, and deviation consistency across the commercial bid, technical bid, and all attachments. A single oversight can mean disqualification. The AI Bid Review Skill packages this checklist-driven methodology into a portable, cross-ecosystem skill that works with any non-scanned PDF or DOCX file and delivers a traceable Markdown risk report.

Under the hood, the skill probes file structure, extracts native text, renders specific pages, and pulls embedded images using Python standard libraries, PyMuPDF, or pypdf. When the underlying model has vision capability, it verifies seals, signatures, table layouts, and designated pages. For scanned documents, image-only Word files, failed conversions, or any point where evidence is insufficient, the skill explicitly flags a skip rather than fabricating a conclusion, so every finding is anchored to a file reference you can audit.

It is built for three audiences: bidding and procurement professionals who want an LLM-assisted first-pass review to save hours of line-by-line comparison; Agent developers on DeepSeek Harness, WorkBuddy, or similar platforms who need a well-bounded, reusable review skill in their workflow; and compliance-sensitive teams that batch-review documents without OCR, network calls, or file uploads. It does not replace formal evaluation, legal opinion, or human final sign-off, and it does not process fully scanned documents.