research-idea-debate SKILL.md skillA
research-idea-debate from daylily-huang/scholarflow is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add daylily-huang/scholarflow/research-idea-debate@git:20260914.c8e6b6bPins git:20260914.c8e6b6b in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- daylily-huang/scholarflow
- Path
skills/research-idea-debate/SKILL.md- Versions recorded
- 4
- Latest version
- git:20260914.c8e6b6b, committed Sep 14, 2026
- First version recorded
- Sep 12, 2026
- Size
- 24,132 bytes
- Repository stars
- 41
- License
- MIT
研究构想与假说推敲技能。从一句模糊直觉或一个既有论断出发，通过用户每轮参与、单焦点提问的多角色讨论，明确概念与比较基准、检查承重前提与证据、推导可观察预测与改判条件、生成并区分竞争解释，最终产出候选研究问题或最小验证方案。内置五类苏格拉底视角（概念澄清者／前提与证据审查者／推论与验证者／替代解释探索者／观点修订引导者）、主持人 P0–P6 动态调度阶梯、想法成熟度守卫（RAW/DEVELOPING/TESTABLE）、关键节点的隔离独立评估与确定性分歧判定、查证缺口先确认后执行的交接规程，以及单写者事件恢复。允许没有任何文献时启动；不代替文献检索、抽取与跨文献综合，需要时按确认范围交接给前三个技能。.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# 研究构想与假说推敲专业技能 (research-idea-debate) 本技能用于在严谨科研场景下，把**模糊直觉**发展成可讨论的研究构想，或对**已有论断**推敲其依据、边界与检验办法（Research Idea Development & Hypothesis Scrutiny）。核心特征是：**用户每轮参与讨论**，而不是观看 AI 自行演完一场辩论。 > **核心哲学**： > 本技能是 **Thinking Partner（思考搭档）**，不是 Judge（裁判），也不是 Summarizer（综述器）。 > 它回答：**"我们可以提出什么解释？这条推理站得住吗？怎样区分和检验？"** > 它绝不回答：**"这个命题对不对？"** —— 经验命题的真伪只能由证据与检验决定。 > > **三条不可让步的底线**： > 1. **多数 Agent 赞成、辩论获胜、没找到反例，都不能证明科学命题正确。** 三者都只是"当前尚未被推翻"。 > 2. **用户每轮参与，AI 不替用户作答、不替用户决定。** 用户说"不知道"时给例子或选项，不给答案。 > 3. **外部意见（导师、同行、会诊）不等于用户决定，也不等于证据。** 它只作为来源明确的一条输入被记录。 --- ## ⚡ 上下文预算与按需渐进加载准则 (Progressive Context Loading Protocol) > [!CAUTION] > **统一执行深度底线规则 (Execution Depth Core Rule)**： > 讨论开始前确认 `execution_depth`；研究目标不能替代深度选择。无有效配置时展示快速／标准／深度三档的轮次上限与范围差异**并只问这一个配置问题**，用户可一句覆盖或回复"按推荐"。未确认不得进入实质讨论。 > **字面确认边界（第二轮实测披露）**：引擎侧深度确认只识别**精确**的"按推荐"（"按推荐吧""就用标准档吧"等变体会被判 unresolved 并重新追问）；上文"授权代选"分支由 Agent 规程执行，不经引擎字面匹配。 > 配置问句（`CONFIGURATION`）不占实质轮次，但不豁免资源计量。 > > **用户拒选／授权代选的分支（实测补充）**：用户回复"你定吧""哪个合适就哪个"这类**授权代选**时，按下列处理，**不得**因此反复追问，也**不得**把它记成用户自选： > 1. 取推荐档（默认标准档），在回复中显式说明"这是按推荐取的，你可随时一句话改成快速或深度"； …
Read the whole file at its exact version.
How to install
mdr add daylily-huang/scholarflow/research-idea-debate@sha256:82eb0fc7b01486efThe line at the top of this page pins the label git:20260914.c8e6b6b, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_f5gqd4tkgmgbebtj)
0 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 4 versions of this file, between Sep 12, 2026 and Sep 14, 2026, more than one a day.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20260914.c8e6b6b latest | 2026-09-14 | c8e6b6b | 24,132 B | A | view · diff |
| git:20260912.bc5555f | 2026-09-12 | bc5555f | 23,501 B | A | view · diff |
| git:20260912.48adfed | 2026-09-12 | 48adfed | 22,698 B | A | view · diff |
| git:20260912.8b2e85d | 2026-09-12 | 8b2e85d | 22,634 B | A | view |
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (24132 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
daylily-huang/scholarflow · 41 stars · license MIT · pushed 2026-11-02 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_f5gqd4tkgmgbebtj GET https://markdownregistry.com/api/v1/resolve?ref=daylily-huang/scholarflow/research-idea-debate GET https://markdownregistry.com/api/v1/blob/82eb0fc7b01486ef2a3c34509f4b951ae2735ac68b20bc98c87e0d99489aae4a
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.