AI Agent Hub
Back to skills
⚖️

Patent Doc Writer

Professional Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Please follow https://skillhub.cn/install/skillhub.md to install @user_0c7f1934/patent-doc-writer into your AI assistant.

About this skill

Problem

Turning a technical solution into a patent specification is often hard because the document must combine legal formatting, claim structure, implementation details, and a clear mapping between method and system claims. Engineers may have a design doc, discussion notes, or requirements, but lack a draft that follows CNIPA conventions.

How It Works

The skill generates the invention patent specification in chapter order. It first collects elements such as invention name, technical field, problem solved, core solution, system modules, and beneficial effects, then drafts the abstract, background, invention content, drawings description, specific implementation, and claims.

  • Claims: It produces method independent claims, method dependent claims, system independent claims, and system dependent claims, with emphasis on characterized-in-that phrasing and correct antecedent references.
  • Implementation: It uses step labels like S101 and S102, and expects concrete parameters such as frame rate, resolution, or thresholds instead of generic preset values.
  • Deliverable: It creates a .docx file using python-docx, followed by a checklist covering abstract length, claim counts, beneficial effects, drawing descriptions, and standard paragraphs.

Boundaries

It is best suited for method and system invention patent drafts in software, electronics, communications, and IoT. The input still needs enough technical detail; vague descriptions like automatic recognition require follow-up on steps, modules, and effects. If the solution processes data, it should also state local processing, privacy safeguards, and security boundaries. It does not replace patent attorney search, office action responses, or legal judgment.

Use Cases

  • A software engineer turns a QR-code recognition design into an invention patent draft with abstract, claims, and implementation steps.
  • A communications algorithm team converts technical discussion notes into a CNIPA-style application draft and checks method-system claim mapping.
  • An IoT product manager completes beneficial effects, drawing descriptions, system modules, and dependent claims for a device-binding solution.
  • A R&D lead compiles system docs into a .docx patent draft and uses the checklist to verify abstract length and claim counts.

Best For

  • Software/algorithm engineers: need a CNIPA-style invention patent draft from a technical solution.
  • Communications/IoT product managers: need system modules, beneficial effects, drawings, and claims organized.
  • R&D leads: need discussion notes or design docs turned into a deliverable .docx with checklist review.
  • Technical writers/patent assistants: need abstract, background, implementation, and claim references completed.