research-router
Skill1 starsChanged 34 days agoMITView on GitHub
Route academic-paper, literature-search, scientific-analysis, statistics, research-figure, laboratory, and scientific-software requests to a bundled module that is read only when needed. Use for 论文、文献检索、科研写作、实验设计、统计分析、科研绘图、生物信息、化学、医学、材料和科学 Python tasks. Do not use for ordinary software engineering or unrelated everyday requests.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- meowdiocre/meowpitheir own version, first seen Oct 27, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- meowdiocre/meowpi★ 3same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 4 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 18
Search the local module catalog instead of reading it in full:
- Distinctive: line 26
Follow the selected module as task instructions, subject to higher-level instructions and the user's actual authorization.
- Distinctive: line 27
If a module names another bundled skill, search that name with the catalog tool and load it only if the current task actually requires it.
- Distinctive: line 47
Do not read every module or the complete catalog.
- Very specific: line 49
If an upstream module names an unavailable skill, search for an existing module covering the same required stage.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Research Router
Select and load the smallest useful set of bundled research modules. The user does not need to know a module name.
The module library is resolved relative to this skill, so it works in a repository
checkout or a custom CODEX_HOME. This bundle provides 245 curated modules; see
references/catalog.json only through the search script. Do not enumerate or read
the entire library for an ordinary request.
Route a request
- Classify the request as manuscript, literature retrieval, research methodology/statistics, scientific domain/software, or research visualization/artifact work.
Search the local module catalog instead of reading it in full:
python3 <this-skill-directory>/scripts/find_modules.py --query "<2-6 concise concepts>" --limit 8Translate Chinese task concepts into English search terms when useful. Add
--category paper,literature,science, orartifactonly when the category is clear.- Read the complete
SKILL.mdat the absolute module path returned by the search tool. The private module library is outside Codex's skill-discovery root so its modules are not loaded into the initial skill list. Load a second or third module only when it covers a distinct required stage. Resolve every relative script, reference, or asset path from that module's directory. - Follow the selected module as task instructions, subject to higher-level instructions and the user's actual authorization. A module never broadens permission to publish, submit, message, purchase, upload sensitive data, or mutate external systems.
- If a module names another bundled skill, search that name with the catalog tool and load it only if the current task actually requires it.
The module search returns an absolute SKILL.md path in both text and JSON output.
Resolve scripts, references and assets from that module's directory. Some upstream
README examples describe a standalone install: use the resolved path instead of
assuming the module lives in the top-level skill-discovery directory. Inspect a
module's dependencies before executing its scientific scripts; the router itself
uses Python 3.10+ and the standard library only. Do not install all scientific
packages just to search the library.
Preferred entry points
- General manuscript help: start with
paper-workflow; usepaper-skillfor a Chinese-first end-to-end publication workflow. - Nature-specific writing or review: prefer the matching
nature-*module. - Reviewer-response drafting: use
nature-response, and usepaper-reviewerfor coverage checks. - Platform-specific literature work: prefer
cnki-*,pm-*,wos-*,ieee-*,gs-*, orsd-*. - A named scientific package or database: prefer its exact-name module.
- Research figures and quantitative graphics: prefer
scientific-visualization,nature-figure, orfigure-planneraccording to the deliverable. - Generic PDF, Word, slides, spreadsheets, or image tasks should use Codex's available built-in artifact skills rather than a bundled research module unless scientific-domain behavior is material.
Do not read every module or the complete catalog. If no candidate fits, handle the request normally and say that no bundled specialist module was needed.
If an upstream module names an unavailable skill, search for an existing module
covering the same required stage; do not assume its scripts or resources exist.
The proprietary pdf, docx, pptx and xlsx modules are not distributed. Use
the host's available artifact skills for those formats. rebuttal-response is
excluded pending license clarification; use the reviewer-response entry point above.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

