onescience-live-literature SKILL.md skillA
onescience-live-literature from onescience-ai/oneskills is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add onescience-ai/oneskills/onescience-live-literature@git:20261030.3665286Pins git:20261030.3665286 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- onescience-ai/oneskills
- Path
skills/onescience-live-literature/SKILL.md- Versions recorded
- 3
- Latest version
- git:20261030.3665286, committed Oct 30, 2026
- First version recorded
- Sep 10, 2026
- Size
- 26,652 bytes
- Repository stars
- 24
- License
- MIT
OneScience 实时文献两级检索推理技能（对齐 Biomni 的 LiteratureSearch/Fetch 工具链），领域无关的通用流程。第一级按相关性实时检索 OpenAlex 拿论文摘要、建全局编号引用池并迭代精炼检索词；第二级对选中的开放获取论文经 Europe PMC/PMC 结构化 API 取全文细节；最终输出分层深度推理答案，References 按正文出现顺序连续编号。触发：任何需要跨论文实时综合推理并带引用列表的问题（物种×性状、通路机制、基因家族、材料体系、药物靶点等）。只产出会话级证据与 harvest 种子清单，不写卡片库。亦可作为 onescience-primitives 本地检索降级到 none/task_only（命中不到可用 Task 知识）时的在线兜底通道被调用；更作为 orchestrator 在步骤 3.6 判定 knowledge_gap=true（本地原语没查到领域实质知识，含空召回 / 仅命中泛化 workflow-planning 流程卡 / domain 不符）时【硬路由兜底】强制调用的在线检索…
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
# OneScience 实时文献两级检索推理 ## 通用性声明（本协议领域无关） - 本技能是**通用流程**，不是针对某一物种/某一主题的专用机制：任何需要跨论文实时综合推理的自然科学问题都走同一套六步协议。 - 协议中**没有任何领域硬编码**：检索词由 Step 1 按问题领域分解产生（物种学名、基因/酶/通路名、方法名、材料名）；全文抽取关键词由 `--keys` 参数传入。 - 换题即换参数，流程与纪律条款不变。 ## 职责边界 - 本技能负责**在线会话级**文献检索与带引用的深度推理回答：摘要层检索 → 迭代精炼 → 全文层取细节 → 深度综合输出。 - 不负责：离线知识卡的撰写/晋升/导出（那是 pkp 流水线的职责）。会话结束时只通过 Step 6 的 seeds.json 交接证据。 ## 协议（六步，严格按序） ### Step 1 查询分解 把用户问题拆成 2~4 组英文检索词（含同义词、物种学名、基因/酶/通路名、方法名）。每组声明 `max_papers`（默认 15）与 `year_from`（默认近 8 年）。同时从问题中提炼 3~6 个**全文抽取关键词**（英文，供 Step 4 的 `--keys` 用）。 ### Step 2 第一级摘要检索（命令块 A） 把下方"命令块 A"模板原样落盘为会话临时脚本 `lit_search.py` 并执行： ``` python lit_search.py "<query>" <max_papers> <year_from> <pool_json路径> ``` - 检索走 OpenAlex works 接口，**相关性排序**（search 默认）+ `from_publication_date` 过滤；禁止改用被引排序（被引排序会漏掉当年新论文，已实测验证）。 - 输出：终端打印编号摘要行；并把条目追加进 pool_json（全局编号跨多次调用续接，形如 [1]~[n]；**按 DOI/标题去重**，重复命中跳过不占编号）。 - 条目字段：title / authors / venue / year / doi / oa_url / cited / abstract（前 400 字）。 ### Step 3 迭代精炼（对齐 Biomni 的第二轮检索） 通读池内摘要，抽出实体词（基因名、酶名、菌株/物种名、通路名、方法名），组成 1~2 组**新**检索词再跑命令块 A，编号续接。至少迭代一轮；若第一轮摘要已出现明确实体名而第二轮未使用，视为协议违规。 …
Read the whole file at its exact version.
How to install
mdr add onescience-ai/oneskills/onescience-live-literature@sha256:372a93f49b9d7dcdThe line at the top of this page pins the label git:20261030.3665286, 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_f6yr6xtll5y3fg5h)
3 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 3 versions of this file, between Sep 10, 2026 and Oct 30, 2026, about one every 25 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20261030.3665286 latest | 2026-10-30 | 3665286 | 26,652 B | A | view · diff |
| git:20260915.3448d36 | 2026-09-15 | 3448d36 | 25,920 B | A | view · diff |
| git:20260910.194f276 | 2026-09-10 | 194f276 | 19,183 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 (26652 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
onescience-ai/oneskills · 24 stars · license MIT · pushed 2026-11-04 · branch master
API
GET https://markdownregistry.com/api/v1/artifacts/art_f6yr6xtll5y3fg5h GET https://markdownregistry.com/api/v1/resolve?ref=onescience-ai/oneskills/onescience-live-literature GET https://markdownregistry.com/api/v1/blob/372a93f49b9d7dcd6675a91235c272641b5ebece2f00c3aba9c02ff6554245e5
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.