/patent-software-ip

name: patent-software-ip
description: 'Generate CN patent docs (claims, specification, abstract) and software copyright materials from AI/big-data
  project code or docs. Covers 7 AI domains + big data, 11 claim templates, auto domain detection, desensitization,
  prior-art search, and self-check. Use when: drafting CN patents from AI code/docs, generating software copyright
  materials, prior-art search for AI inventions, 专利撰写/软件著作权/AI知识产权/脱敏处理.'
license: Apache-2.0
metadata:
  version: 1.4.1
  author: jaccen
  tags:
  - patent
  - software-copyright
  - ip
  - ai
  - big-data
  - 3d-vision
  - generative-ai
  - embodied-ai
  - nlp
  - rag
  - ai-engineering
  - ai-safety
  when_to_use:
  - Draft CN invention patents from AI code or docs
  - Generate software copyright materials
  - Perform prior-art search for AI inventions
  - Desensitize project code for IP submission
  - Auto-detect AI domain and select claim template
  - Self-check patent specification completeness
  - 专利撰写 / 软件著作权 / AI知识产权 / 脱敏处理 / 现有技术检索
  platform-fields:
    user-invocable: true

Patent & Software Copyright Generation (AI + Big Data)

Generate CNIPA invention patent documents or CPCC software copyright materials from AI / big-data project code, design docs, and research papers.

**Covers 7 AI domains + Big Data (23 sub-directions), 11 claim templates.**

Full version (Chinese, with Word/PPT output): see `AI-Copyright-Skill` project.

Triggers

patent / claims / specification / software copyright / disclosure / IP application / paper-to-patent / `/patent-software-ip`

Overall Flow

Phase A  Requirement Diagnosis -> path + domain classification + risk level
Phase B  Project Analysis -> auto-detect domain + extract key technical points
Phase C  Generation (branch by path)
  C1 Patent: prior art search -> claims (11 templates) -> specification -> abstract -> self-check
  C2 Software Copyright: manual -> source code doc -> self-check
Phase D  Iterative Correction

Phase A: Requirement Diagnosis

Confirm: path (patent/copyright/both), tech topic, applicant/inventor info, existing materials.

**Auto domain classification** (see Section "AI Domain Taxonomy" below).

**Gate**: 3-5 line diagnosis summary including domain + risk level.

AI Domain Taxonomy

| Domain | Sub-directions | High-Risk Flags | |--------|---------------|-----------------| | **D1 Perceptual Intelligence** | 2D vision, 3D vision, multi-sensor fusion | 3D vision: bind 4-stage pipeline | | **D2 Cognition & Language** | NLP, multimodal LLM, RAG, knowledge graph | RAG: show full 5-stage chain | | **D3 Generative AI** | Diffusion, LLM text gen, cross-modal gen, AIGC watermark | **Must bind condition injection method**; pure content gen = rejected | | **D4 Decision & Interaction** | Embodied AI, reinforcement learning, multi-agent | **Must bind sensor + actuator**; RL: bind reward to concrete task | | **D5 AI Engineering** | Training/fine-tuning, inference deployment, data engineering, edge IoT | Training: bind to specific model architecture; inference: bind to hardware | | **D6 AI Safety & Governance** | Adversarial robustness, watermark/tracing, privacy, alignment, 3DGS provenance | Need concrete technical measure, not policy-level description. 3DGS IP: reference GaussTrace (ICML 2026) for evidence-driven provenance graphs | | **D7 Industry Applications** | Autonomous driving, industrial, medical, financial, AI4Science | **Must bind data processing means**; financial: bind to data analysis | | **D8 Big Data** | Distributed computing, data pipeline, stream processing, data quality, real-time analytics | **Must bind to specific application scenario**; pure platform = rejected |

Phase B: Project Analysis

B.1 Auto-Detection Decision Tree

Source files -> domain mapping:

| Key file | Detected domain | |----------|----------------| | `model.py`, `unet.py`, `vae.py` | D3 Generative AI | | `train.py`, `finetune.py` | D5 AI Engineering (Training) | | `inference.py`, `triton_serve.py`, `onnx_export.py` | D5 AI Engineering (Inference) | | `render.py`, `gaussian.py`, `splat.py` | D1 3D Vision | | `llm.py`, `chat.py`, `rag_chain.py` | D2 NLP / RAG | | `robot.py`, `vla.py`, `env.py` | D4 Embodied AI | | `reward.py`, `ppo.py` | D4 Reinforcement Learning | | `watermark.py`, `embed_watermark.py` | D6 AI Safety / Watermark | | `spark_job.py`, `flink_job.py`, `kafka_consumer.py` | D8 Big Data | | `etl.py`, `data_pipeline.py`, `feature_store.py` | D8 Big Data (Data Engineering) | | `stream.py`, `realtime_analytics.py` | D8 Big Data (Streaming) | | `dataset.py`, `dataloader.py` | D5 AI Engineering (Data) | | `privacy.py`, `dp_train.py` | D6 AI Safety (Privacy) | | `config.yaml`, `pipeline.py` + langchain | D2 RAG / Agent |

Also detect 6 industry contexts: medical, financial, autonomous driving, industrial, smart city, education.

B.2 Technical Points Extraction

Priority: model definition -> training/inference -> domain-specific core -> papers/design docs -> README.

Output: **Key Points List** (innovations, scheme skeleton, key params, distinctions, quantifiable effects, domain classification).

**Gate**: Present key points list for user confirmation.

Phase C1: Patent Application

C1.1 Prior Art Search

Online search 2-3 rounds: CNIPA patent DB, Google Patents, arXiv. Each result: source ID, scheme summary, limitations.

**CPC suggestions by domain**:

• D1 3D Vision: G06T 7/50, G06T 17/00
• D2 NLP/RAG: G06F 40/30, G06N 3/08
• D3 Generative AI: G06N 3/045, G06T 13/00
• D4 Embodied: G05B 19/00, B25J 9/16
• D5 AI Engineering: G06N 3/084
• D6 AI Safety: G06F 21/60
• D7 Industry: varies by sector
• D8 Big Data: G06F 16/245, G06F 16/903

C1.2 Claims (11 Templates)

**Structure**: Method (1 independent + 3-8 dependent) + System (1 independent + 3-8 dependent) + Storage Medium (1 independent).

**Template selection by domain**:

| Template | Domain | Independent claim skeleton | |----------|--------|--------------------------