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 Liuziyu77/gene-skill --skill research-writer-combinedgit clone --depth 1 https://github.com/Liuziyu77/gene-skillWrote 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/liuziyu77/gene-skill/research-writer-combined)<a href="https://agentmods.dev/skills/liuziyu77/gene-skill/research-writer-combined"><img src="https://agentmods.dev/badge/skills/liuziyu77/gene-skill/research-writer-combined/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/liuziyu77/gene-skill/research-writer-combined"><img src="https://agentmods.dev/badge/skills/liuziyu77/gene-skill/research-writer-combined.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00139 | $0.01792 |
| Opus 5 | $0.00069 | $0.00896 |
| Sonnet 5 | $0.00028 | $0.00358 |
| Haiku 4.5 | $0.00014 | $0.00179 |
Grade A, and why
research-writer-combined 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 13d 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 — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.
调研-写作 合成引擎
大多数人的流程是：先调研完，再开始写。 这个合成体的流程是：边写边调研——每一个论断触发一次针对性搜索，每一次搜索直接喂进草稿。
触发条件
- 需要产出「有来源支撑的深度内容」（分析报告、研究文章、带论据的提案）
- 用户有主题和初步观点，但需要事实支撑和结构化
- 需要快速出一份「可交付」的内容，而非单独调研报告或单独草稿
核心工作流（T1）⚡超级基因：论据优先写作流
来自：huashu-research（T1-分层调研）× structured-writer（T1-大纲驱动写作）协同涌现
传统流程的问题：
- 先全面调研：收集了大量素材，写作时发现 80% 用不上，浪费
- 先列大纲写作：写到需要数据支撑的地方才去搜索，写作节奏被打断
合成体的论据优先写作流：
Phase 0：主题锚定（5 分钟）
确认三件事：
- 核心论点（不超过一句话，这篇内容要说服读者相信什么）
- 目标读者（决定深度、术语密度、论证方式）
- 交付格式（几千字？几个 section？有没有引用格式要求）
Phase 1：骨架优先写作
先写出无来源版草稿骨架（标题 + 每段第一句话）：
- 不停下来搜索
- 遇到「这里需要数据/案例」，用
[需要：XXX]占位 - 目标：10 分钟内完成全文骨架
Phase 2：针对性调研（并行）
扫描所有 [需要：XXX] 占位符，按重要性排序，并行启动针对性搜索：
- 每个搜索任务只解决一个
[需要：XXX] - 找到后立即填入草稿对应位置
- 来源质量判断标准（来自 huashu-research）：一手 > 权威媒体 > 二手转述
Phase 3：扩写与润色
骨架填充完成后，逐段扩写：
- structured-writer 的节奏控制：每段「论点句 → 证据 → 分析 → 过渡」四件套
- huashu-research 的质量把关：每个核心论断必须有来源，推断必须标注「作者判断」
Phase 4：结构验证
逐段检查：
- 论点是否被论据支撑（而不是反过来，论据决定论点）
- 相邻段落是否有逻辑连接（不是简单罗列）
- 结尾是否回应了开头的核心论点
工具调用模式（T2）
| 场景 | 工具 |
|---|---|
| Phase 0 快速背景扫描 | WebSearch（2-3 次，建立认知地图） |
| Phase 2 针对性搜索 | WebSearch × N（每个占位符一次） |
| 精确读取重要文章 | web-article-reader（如有） |
| PDF/书籍素材 | Read |
| 输出写入文件 | Write |
工具节约原则（来自 huashu-research）：宁可少搜几次但每次质量高，不要广撒网收集噪音。
输出格式（T3）
标准交付物（Markdown）：
# [标题]
> [核心论点：一句话]
## 一、[Section 1 标题]
[正文，每段：论点+证据+分析]
## 二、[Section 2 标题]
...
---
**参考来源**
- [来源1](URL) — [一句话说明用途]
- [来源2](URL) — [一句话说明用途]
**作者判断标注**：本文 [X] 处包含作者未经来源支撑的推断，已标注「作者判断」。
交互模式（T4）
| 检查点 | 内容 |
|---|---|
| Phase 0 结束 | 确认核心论点 + 目标读者 + 格式 |
| Phase 1 结束 | 展示骨架，用户可调整结构再进 Phase 2 |
| Phase 2 结束 | 告知找到/未找到的占位符，未找到的处理方案 |
| Phase 4 结束 | 交付前最终确认，用户可要求调整任意段落 |
错误处理（T5）
| 错误类型 | 处理 |
|---|---|
某 [需要：XXX] 找不到可靠来源 |
标注「来源缺失」，在结尾列出；或改为「作者判断」明确说明 |
| 搜索结果与核心论点矛盾 | 不回避——在正文中呈现矛盾，修正论点或增加「反驳与回应」段落 |
| 用户要求「加快」（跳过调研） | 提示风险：哪些论断将变为无来源推断，用户确认后继续 |
| 字数超出限制 | 先完成结构，再按重要性裁剪证据，保留论点不打折 |
内在张力（G5）
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.
- 13d ago First seen · 161 lines · 139 tokens per session scan A 4429cb11f0de
research-writer-combined is a skill published in the GitHub repository Liuziyu77/gene-skill (56 stars, last pushed 4mo ago), licensed MIT. It adds 139 tokens to every session and 1,792 once invoked, about $0.0007 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
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…