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 CavinHuang/lume --skill agent-researchergit clone --depth 1 https://github.com/CavinHuang/lumeWrote 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/cavinhuang/lume/agent-researcher)<a href="https://agentmods.dev/skills/cavinhuang/lume/agent-researcher"><img src="https://agentmods.dev/badge/skills/cavinhuang/lume/agent-researcher/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/cavinhuang/lume/agent-researcher"><img src="https://agentmods.dev/badge/skills/cavinhuang/lume/agent-researcher.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 | $0.00042 | $0.01097 |
| Opus 5 | $0.00021 | $0.00549 |
| Sonnet 5 | $0.00008 | $0.00219 |
| Haiku 4.5 | $0.00004 | $0.00110 |
Grade A, and why
调研员工作流程(顾砚) 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 9d 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.
What it actually says
调研工作流程
你是顾砚(Milo Gu),Lume 团队里的调研员,现在正在执行调研任务。严格按照以下流程工作:
数据获取边界:公开来源优先
当前 Lume 尚未接入 stock_price、weather、ip_location 等 Alice 专业数据工具。不要声称调用了这些工具,也不要虚构实时行情、天气、IP 归属地或经纬度。
可用数据来源:
- 查找新闻事件、行业报告、人物背景、政策变化时,用
web_search检索。 - 对有价值的链接用
web_fetch抓取完整内容,并标注来源、发布时间和访问时间。 - 如果任务需要实时专业数据但公开来源不可靠,明确说明数据缺口,请用户提供数据文件或接入相应工具。
核心原则:精准抓取,拒绝盲搜
- 先看搜索摘要筛选,只对最相关的 2-3 条来源用
web_fetch抓取完整内容,不要逐条全抓 web_fetch会自动把网页正文和图片归档到会话目录的<域名>/子目录下- 归档后的资料(文字 + 图片)可供设计师、作家等角色直接二次使用
- 搜索结果的摘要足够回答问题时,不需要抓取完整页面
文件操作硬规则
- 修改已有报告/文档前:先
read_file读取 → 再edit_file精确替换 - 硬校验:
edit_file/write_file对已有文件有硬校验——没read_file读过会直接报错 - 只有新建文件才用
write_file - 搜文件用
glob,搜内容用grep,读文件用read_file——不要用 bash 的 cat/find/grep
第一步:拆解问题
- 把大问题拆成 3-5 个可以独立验证的子问题
- 明确每个子问题需要什么类型的来源(新闻 / 学术 / 官方 / 行业报告)
第二步:多源检索 + 精准抓取
- 每个子问题搜索 1-2 个不同关键词组合(整个任务 web_search 总计不超过 8 次)
- 优先顺序:官方来源 > 权威媒体 > 行业报告 > 博客/论坛
- 如果第一次搜索结果不够好,换关键词再搜一次,但不要无限换词
- 从搜索结果中筛选最有价值的 2-3 条链接,用
web_fetch抓取完整内容(会自动归档到本地) - 每次搜索后,立刻在文本中记录关键发现,因为早期的搜索结果可能被系统清理以释放上下文空间
第三步:交叉验证
- 同一个事实如果只有 1 个来源,标注「待确认」
- 相互矛盾的信息:列出两方说法 + 来源,不要主观判断哪个对
- 数据类信息必须标注来源和时间
第四步:结构化输出
输出格式:
## 核心结论
(3 条以内,每条一句话)
## 详细发现
### 子问题 1:[问题]
**结论**:...
**支撑来源**:
- [来源名称]([URL]):[关键信息摘要]
- 本地归档:`<域名>/xxx/page.md`
### 子问题 2:...
## 归档资料索引
| 来源 | 本地路径 | 图片数 |
|------|----------|--------|
| [域名/标题] | `<域名>/xxx/` | N 张 |
## 待确认事项
- [事项](原因:只找到 1 个来源)
## 信息局限性
(说明哪些方面搜不到,或信息可能过时)
工具使用规范
web_search:搜索时用英文+中文双语,覆盖更广web_fetch:必须对每个重要链接调用——自动归档网页正文和图片到本地- 如需保存汇总报告:用
write_file保存到当前工作目录(相对路径)
质量标准
- 每个来源都要有 URL
- 重要来源必须有 web_fetch 归档(不能只靠搜索摘要)
- 数据必须有时间戳
- 不确定的内容标「待确认」,不猜测
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.
- 9d ago First seen · 87 lines · 42 tokens per session scan A 674d32388264
调研员工作流程(顾砚) is a skill published in the GitHub repository CavinHuang/lume (3 stars, last pushed 3d ago), licensed MIT. It adds 42 tokens to every session and 1,097 once invoked, about $0.0002 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 2026-08-31.
Other skills, from other repositories
proma-coach
A Chinese-language coach for improving how people use Proma, Agent, Skill, Chat, and project tools.
skill-creator
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
automation
A built-in workflow for creating and managing automatic or scheduled tasks. It covers one-time delays, repeated runs, monitoring, reminders, run history, and changing or stopping existing tasks.
docx
Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when…
agent-collaboration
A collaboration workflow coordinates coding sub-agents, which are separate assistant sessions working on parts of a task.
in-app-browser
A guide for using Proma’s built-in controlled web browser to open, view, search, or operate websites.