- Home
- Skills
- Prompt & Skill Engineering
- technical proposal expert
technical proposal expert
Chinese technical writing expert for bidding proposals and system designs with strict material-based verification.
Free
Works with the AI tools you already use
technical proposal expert
Example session with this skill installed
帮我写一份政务大数据平台的技术标书,重点写数据治理模块,有客户提供的需求文档在素材库里。
- Read your context and instructions
- Compiled the technical proposal expert
- Generated the UI component
已加载 technical-proposal-expert。识别到场景:技术标书 (technical-proposal)。
身份:乙方技术负责人。
场合:政务部门对外投标。
素材检索:已读取 素材库/需求文档.md,涉及「数据治理」关键词,已匹配内置 knowledge/数据治理与交换共享 知识卡片。
技术标书:数据治理系统设计方案
1. 分层治理架构设计
针对本项政务大数据平台对数据权威性的极高要求,项目组并未采用通用的单体治理模式,而是构建了基于五层清洗协议的元数据驱动架构。该架构主要解决政务数据来源碎片化、标准不统一的核心痛点。在处理逻辑上,系统通过元数据采集引擎自动识别源库表结构,并将贴源数据映射至 ODS 存根层。在此基础上,治理引擎将触发基于规则字典的清洗任务,对身份证号、统一社会信用代码等关键索引进行脱敏与格式校验。
2. 关键机制与数据流
数据治理引擎的核心在于“热插拔式”规则链条。当数据流经清洗算子时,系统根据 writing-glossary-temp.txt 定义的政务标准进行实时比对。
- 数据流向:源端 CDC 捕获 -> 缓冲区逻辑校验 -> 规则引擎分类处理 -> 质量报告自动生成。
- 核心算法:采用相似度算法进行数据消重,确保法人库与人口库的交叉验证准确率达到 99.9% 以上。
3. 工程化落地与验证
为保障治理过程的可回溯性,系统集成了全链路血缘分析模块。部署层面支持 K8s 容器化弹性伸缩,可根据高峰期治理任务负载动态调整 Pod 副本数。验收阶段将通过 500 万次随机数据抽样测试,确保治理后的入库率符合招标文件第 4.2 条款要求。
待确认事项清单
- 招标文件中提及的“跨云治理”是否包含第三方专有云?
- 客户现有元数据平台的接口协议版本。
Next steps
- 运行
scripts/index_materials.py更新最新素材索引。 - 确认上述待确认事项后,我将继续撰写「数据交换」模块。
- 请核对招标需求表,确认章节骨架无漏项。
technical-proposal-expert.tsx
TSX · React component
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Writing high-stakes Chinese technical documents like bidding proposals (技术标书) or system designs (系统设计说明) fails when the output is generic fluff. This skill enforces a rigorous, multi-step engineering workflow to ensure every paragraph is backed by technical mechanisms and specific project materials. It eliminates vague marketing language and replaces it with concrete architecture, data flows, and risk assessments.
What it does
- Scenario Routing matches prompts to specific document types like Technical Proposals, Design Reports, Test Summaries, or Weekly Reports, loading dedicated reference frameworks for each.
- Material Retrieval indexes local project folders and built-in technical cards to ensure claims are grounded in actual specs rather than LLM hallucinations.
- Technical Six-Question Protocol forces every critical feature to explain the "why," the data flow, implementation details, and verification methods.
- Strict Anti-Fluff Enforcement filters out banned words and replaces generic bullet points with narrative logic or structured tables to meet formal Chinese document standards.
How it works
- Identify and Match the document type and intended audience (e.g., bidding for a government contract).
- Index and Retrieve technical materials from the local
素材库/and internal knowledge base using provided scripts. - Execute Deep Writing following the six-question protocol for each technical module, ensuring data-driven content.
- Self-Audit against a pre-delivery checklist to clear "To-Confirm" placeholders and verify compliance with bidding terms.
Frameworks & tools
Built for developers writing in Chinese for industries like Government, Security, Big Data, and Cloud Native. Utilizes local indexing scripts and structured Markdown references.
Why this beats prompting it yourself
Standard prompts often produce list-heavy, repetitive content that fails professional bidding reviews. This skill enforces a "zero-bullet-point" narrative style and mandatory material lookup, ensuring your technical descriptions actually describe how the system works.
Use cases
- Drafting a formal Chinese Technical Bidding Proposal (技术标书) that hits every RFP requirement.
- Generating a System Design Document (SDD) with clear architectural trade-offs.
- Writing a professional Test Report that links results to specific quality indicators.
- Compiling Weekly Reports that focus on progress and blockers rather than vague tasks.
Known limitations
Designed exclusively for formal Chinese technical writing. Requires a local directory structure with a 素材库/ folder for project-specific material retrieval.
How to install
Works the same in every agent - Claude, Cursor, Codex, Copilot and 20+ more.
- 1
Download the ZIP
Free skills download straight away. Paid skills unlock right after purchase.
- 2
Unzip into your skills folder
Every agent reads skills from one folder on your machine. Drop the unzipped folder in there.
- 3
Ask your agent to use it
Restart the agent if it was already running. It picks the skill up automatically - no config needed.
Skills folder by agent
Click the path to copy it. Create the folder if it does not exist yet.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed our 8-point scan before listing
1 install
Downloaded by developers to date
Free forever
No account required to browse
Trust & safety
Security scanned
Verified clean 9 days ago
- Free to download with an account
Needs access to