Bear Scoop logo

Bear Scoop

Community
fei0810
bear-scoop

在你投入一个研究方向之前,先看看有谁已经在做同样的事。把 idea 拆成多个不同角度分别去检索,返回一张"邻近程度地图",重点标出最危险的撞车候选。底层走真实检索(scimaster-cli),不做主观新颖性判断。 **以下情况请主动触发本技能**:用户说"这个选题有没有人做过"、"我会不会被抢发"、"这个方向还有空间吗"、"我的 idea 新不新"、"要不要查一下有没有撞车"、"这个想法是不是已经有人发了"——即使用户没有说"bear-scoop",只要意图是**在提交或动笔前确认 idea 的独特性**,就使用本技能。加 `-r` flag 只看近两年和预印本,适合担心被抢发的场景。 不适用于:为已有观点找支持文献(用 bear-support)、找反对文献(用 bear-counter)、概念地图(用 bear-map)、溯源演化史(用 bear-trace)。

Overview

Publisherfei0810
Repositorybear-research-skills
Skill namebear-scoop
Stars
73
Forks
16
Bundled files
2
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 2 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by fei0810 on GitHub. Read the source before you install it.

Installation

Install the Bear Scoop AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/fei0810/bear-research-skills/tree/main/skills/bear-scoop
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/fei0810/bear-research-skills.git /tmp/bear-research-skills
mkdir -p .claude/skills
cp -r /tmp/bear-research-skills/skills/bear-scoop .claude/skills/bear-scoop
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Bear Scoop in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Bear Scoop on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Bear Scoop is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

bear-scoop · 选题撞车检测

一个任务:在你投入之前,看看谁坐在离你最近的位置。抢发你的那组人,用的词往往不是你会用的词——这里从你不会想到的角度去搜。

先读 references/sci-cli.md——CLI 检测步骤、用法和铁律。 再读 references/output-system.md——三层输出体系和 HTML 外壳规范。

Flags

  • 默认:全角度扫描
  • -r:只看近一两年和预印本(撞车风险最高的地方)

步骤(固定顺序)

Step 0 — 确认 CLI 可用 运行 sci --version。未安装就给安装命令并停止;认证错误就提示 sci init 并停止。

Step 1 — 压缩 idea 把 idea 提炼成一句话并确认。模糊的 idea 搜出来的是噪音,不是信号。

Step 2 — 分解成 5–6 个正交角度 同一项工作,从不同立场描述:

  1. 字面表述(用户自己的说法)
  2. 方法中心(技术 / 工具,剥离应用场景)
  3. 问题中心(要解决的问题,剥离方法)
  4. 结论中心(预期会声称的具体结果)
  5. 相邻领域的表达方式(另一个学科会怎么叫这件事)
  6. 抢发者的标题(如果有人比你快发了,那篇论文会叫什么)

在输出里明确列出这 5–6 个角度,每个附一行说明和实际使用的查询词。 用户需要看见你从哪些方向搜了,才能判断覆盖是否够全,也能自己补充遗漏的角度。列完后直接开始检索,不等待用户确认(除非 idea 本身太模糊)。

Step 3 — 分角度检索 宽泛角度用 ultra_low(免费),有收获的两三个用 low 复跑。加 -r 时偏向近期结果。每个角度用独立的 --prefix。

Step 4 — 合并去重 按 DOI 或标题去重。

Step 5 — 按邻近程度分四层

  • 直接撞车:问题 + 方法 + 结论都重叠
  • 方法孪生:你的方法,别人的问题
  • 问题孪生:你的问题,别人的方法
  • 邻居:相关但不重叠

每篇附一行"为什么它在这一层"。最危险的那篇单独在最上方点名。

分层前先给每篇候选做四轴评分,每轴 0 到 2 分:

  1. 问题重叠:是否解决同一个问题
  2. 方法重叠:是否使用同一技术路线
  3. 结论重叠:是否会声称相同或高度相似的发现
  4. 场景重叠:人群、数据、材料、任务或应用场景是否接近

分层规则:

  1. 直接撞车:总分 7 到 8,且问题、方法、结论三轴都不低于 1
  2. 方法孪生:方法 2 分,问题或场景明显不同
  3. 问题孪生:问题 2 分,方法明显不同
  4. 邻居:总分 1 到 3,或只共享背景领域

如果评分和直觉冲突,按评分分层,并在"为什么它在这一层"里解释冲突。

输出格式

搜索角度列表后,直接给撞车地图(★ 是你的 idea,从危险到安全向外展开):

── 撞车地图 ──────────────── [你的 idea] ──

[!] 直接撞车
    Smith 2023 — "论文标题"
    邻近度  ████████░░

    方法孪生
    Jones 2022 — "论文标题"
    邻近度  ██████░░░░

    问题孪生
    Wang 2021 — "论文标题"
    邻近度  ████░░░░░░

    邻居
    Li 2020 — "论文标题"
    邻近度  ██░░░░░░░░

──────────────────────────────────────────

没有文献的层:— [层名] 此层未检索到,不从地图消失。 结尾两句:哪里拥挤,哪里安静。

输出物

检索结束后创建 {topic-slug}/ 子文件夹,输出四个东西:

  1. 终端 — 直接 print TUI 格式(撞车地图 ASCII)
  2. report.md — 按 output-system.md 的结构化 Markdown 规范生成:稳定 YAML front matter + 一眼结论 + 撞车分层 + 选题动作建议 + 证据表 + 详细证据 + 检索透明度 + 可复用 JSON
  3. report.html — 自包含 HTML,按 output-system.md 规范生成
  4. references.bib — 合并所有 query 生成的 .bib 文件,去重后只保留报告中实际引用的文献条目

HTML 签名体验:SVG 撞车雷达 + 四层文献 tabs,强调色 #2563eb(蓝)。雷达悬停 tooltip 用 JS mousemove 实现(不用 <title>)。HTML 组件规范见 output-system.md 的 bear-scoop 节。

简单查询只给 1–2 条轻量下一步。开题或担心抢发时给具体建议:是否继续推进、该避开哪个点、题目怎么改。

诚实边界

没有检索到直接撞车,不等于你是新的。可能是刚刚发出的预印本、付费墙后面的文章、未收录的期刊、或者根本还没人投稿。报告检索到的内容和它说明了什么——不要说"你是安全的"。完整的新颖性论证是产品级任务,交给 SciMaster。

Gotchas

  • 雷达图点位不能为空:用 CSS div 同心圆时无法在圆环上放置点位,必须用 SVG。空圆环对用户没有任何信息价值。
  • 四轴评分不要过于严格:问题/方法/结论/场景四轴各 0-2 分,总分 8 分。现实中 7 分和 8 分的文献对用户来说撞车风险相同,不要因为一轴差 1 分就把论文从"直接撞车"降到"方法孪生",这会低估真实风险。
  • 综述类文献不要放进直接撞车:综述和原创研究的竞争性质不同。综述几乎不构成"你的选题被抢发"的威胁,应归入"邻居"层。
  • 不要因为找不到直接撞车就报告"安全":数据库覆盖有限,这个结论没有依据。

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Bear Scoop AI skill do?

在你投入一个研究方向之前,先看看有谁已经在做同样的事。把 idea 拆成多个不同角度分别去检索,返回一张"邻近程度地图",重点标出最危险的撞车候选。底层走真实检索(scimaster-cli),不做主观新颖性判断。 **以下情况请主动触发本技能**:用户说"这个选题有没有人做过"、"我会不会被抢发"、"这个方向还有空间吗"、"我的 idea 新不新"、"要不要查一下有没有撞车"、"这个想法是不是已经有人发了"——即使用户没有说"bear-scoop",只要意图是**在提交或动笔前确认 idea 的独特性**,就使用本技能。加 `-r` flag 只看近两年和预印本,适合担心被抢发的场景。 不适用于:为已有观点找支持文献(用 bear-support)、找反对文献(用 bear-counter)、概念地图(用 bear-map)、溯源演化史(用 bear-trace)。

Why use Bear Scoop on TypingMind?

Because you install it once and use it with any model. Bear Scoop is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Bear Scoop in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/fei0810/bear-research-skills/tree/main/skills/bear-scoop. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Bear Scoop?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Bear Scoop?

As many as you like. As long as a model supports skills, you can use Bear Scoop with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Bear Scoop AI skill free?

It is published on GitHub by fei0810. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇