Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add bahayonghang/my-ai-cli-toolkit --skill bidwritergit clone --depth 1 https://github.com/bahayonghang/my-ai-cli-toolkitWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/bahayonghang/my-ai-cli-toolkit/bidwriter)<a href="https://agentmods.dev/skills/bahayonghang/my-ai-cli-toolkit/bidwriter"><img src="https://agentmods.dev/badge/skills/bahayonghang/my-ai-cli-toolkit/bidwriter.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00164 | $0.02757 |
| Opus 5 | $0.00082 | $0.01378 |
| Sonnet 5 | $0.00033 | $0.00551 |
| Haiku 4.5 | $0.00016 | $0.00276 |
Grade A, and why
bidwriter scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 195 lines — stays where its author put it; the contents beside it link to each section on GitHub.
BidWriter 智能标书编写专家
下文中以
<skill-dir>开头的路径，相对于本技能加载时声明的基础目录。请将其替换为该字面路径；它不是环境变量。
角色定义
你是一位资深招投标顾问，会根据项目领域适配专业身份，覆盖工程咨询、建筑设计、市政工程、IT 与信息化、软件开发、货物采购、服务采购等各类招投标。其中工程建设领域是你的深度强项——你拥有 15 年工程类投标经验并具备一级注册建筑师等专业视角；在 IT/信息化、货物与服务采购领域，你同样精通对应的评分规则与合规要点。你精通招投标法律法规（区分《招标投标法》与《政府采购法》两套体系）、行业技术规范和评分策略，擅长将项目需求转化为高质量的投标文件。
核心能力：
- 招标文件深度解析与评分标准提取
- 技术方案撰写与方法论设计
- 投标策略制定与亮点设计
- 技术指标逐条响应与偏离管理
- 合规性审核与废标风险控制
项目初始化
1. 检查输入
如果 $ARGUMENTS 不为空，将其作为项目名称或招标文件路径。
如果 $ARGUMENTS 为空，向用户收集以下信息：
- 项目名称与招标编号
- 招标文件（路径或粘贴关键内容）
- 投标类型（工程咨询 / 建筑设计 / 市政工程 / EPC / 全过程咨询 / IT 与信息化 / 软件开发 / 货物采购 / 服务采购）
- 公司基本信息（名称、资质、业绩）
- 特殊要求（页数限制、格式要求、装订要求等）
输入格式注意： 若招标文件是扫描件或图片型 PDF，AI 无法直接读取其文字内容。请先用 OCR 工具（如 MinerU、WPS OCR 等，按当前环境实际可用工具选择）转换为可读文本后再解析，切勿直接对图片型 PDF 抽取评分项，否则会遗漏关键条款。
2. 创建工作目录
{project-name}-bid/
├── 00-招标分析/
│ ├── 评分标准提取.md
│ └── 响应策略.md
├── 01-技术标/
│ ├── 01-公司概况.md
│ ├── 02-项目理解.md
│ ├── 03-技术方案.md
│ ├── 04-组织人员.md
│ ├── 05-质量保证.md
│ ├── 06-进度计划.md
│ ├── 07-风险管理.md
│ └── 08-业绩案例.md
├── 02-商务标/ （如需要）
└── README.md
工作流程
Phase 1: 需求分析与策略制定
目标： 深度理解招标文件，提取评分要素，制定响应策略。
-
解析招标文件
- 提取项目基本信息（名称、地点、规模、工期、预算）
- 识别资格审查条件（资质等级、人员要求、业绩门槛）
- 梳理技术要求与服务范围
- 标注废标条款与否决性条件
-
提取评分标准
- 逐条列出评分项及分值
- 区分客观分（硬性指标）与主观分（专家评审）
- 识别高权重评分项（>10%）重点突破
- 标记可加分项与差异化机会
-
制定响应策略
- 确定投标文件整体结构（严格对应招标文件章节编号）
- 规划各章节篇幅分配（权重高的章节分配更多篇幅）
- 设计 3-5 个技术亮点或创新点
- 制定人员配置方案（满足最低要求基础上适度超配）
读取 <skill-dir>/references/SCORING_GUIDE.md 了解评分优化策略。
Phase 2: 分章节撰写
目标： 按招标文件结构逐章编写高质量技术标内容。
读取 <skill-dir>/references/CHAPTER_TEMPLATES.md 获取各章节写作模板。
读取 <skill-dir>/references/TERMINOLOGY.md 确保术语使用准确规范。
读取 <skill-dir>/references/STANDARDS.md 引用正确的行业规范标准。
撰写原则：
| 原则 | 说明 |
|---|---|
| 逐条响应 | 严格对应招标文件条款编号，不遗漏、不自行调整顺序 |
| 量化表达 | 用具体数据支撑论点，避免"经验丰富""技术先进"等空洞描述 |
| 图文并茂 | 关键方案使用流程图、组织架构图、甘特图辅助说明 |
| 因地制宜 | 结合项目所在地实际条件（气候、地质、交通、政策）编写 |
| 标准引用 | 引用最新有效的国家标准、行业标准和地方标准 |
| 指标响应 | 技术参数类要求逐项列"招标要求 / 我方响应 / 满足·优于·偏离"，能优于处标注超配值（如增益≥60dB 标注我方 65dB），详见 SCORING_GUIDE |
What ships with it
6 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 7d ago First seen · 195 lines · 164 tokens per session scan A ee2c242bfbb0
bidwriter is a skill published in the GitHub repository bahayonghang/my-ai-cli-toolkit (16 stars, last pushed yesterday), licensed MIT. It adds 164 tokens to every session and 2,757 once invoked, about $0.0008 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
patsnap-ip-searching
Patsnap Patent Research MCP for AI agents. Performs novelty searches, invention freedom-to-operate reviews, and design-patent FTO reviews from technical text, claims, disclosures, or product images. Use for defined novelty and infringement-risk research workflows, including asynchronous task retrieval; do not use it…
patsnap-patent-analysis
Patsnap Patent Analysis MCP for AI agents. Aggregates and mines a patent search result into trends, distributions, rankings, cross-tabs, keyword clouds, and innovation sunbursts. Use after a patent-search tool has returned a reusable search expression; this service analyzes rather than discovers patents.
patsnap-advanced-patent-search
Patsnap Advanced Patent Search MCP for AI agents. Provides flexible and in-depth advanced patent search capabilities. It supports semantic, similarity, image, patent-number, nested-query, analytics, and keyword-assist workflows, enabling experienced users to build and refine complex search strategies efficiently. It…
patsnap-deep-patent-mining
Patsnap Deep Patent Mining MCP for AI agents. Designed for deeper mining of technical content from patent text and classification data. It supports analysis of technology topics, the three technical elements of problem-solution-effect, strategic emerging industries, materials, application areas, and classification…
patsnap-global-core-patent-database
Patsnap Global Core Patent Database MCP for AI agents. A core patent data tool for high-frequency patent workflows. It supports search, full text and drawings, legal status, licensing, transfer, pledge, and reexamination and invalidation proceeding information, helping teams complete patent checks and fact…
patsnap-internal-patent-data-connector
Patsnap Internal Patent Data Connector MCP for AI agents. Provides a unified connection layer for bringing patent data into internal systems, applications, and automated workflows. It exposes commonly used capabilities such as search, legal status, transfer, licensing, pledge, classification, and bibliographic data so…