菜单
Convert normalized multi-search-engine results to minimal BibTeX, preserving real DOI, author, publication year, arXiv ID, source URL, and provenance metadata when available. Unknown years are omitted and duplicate DOI records collapse to one entry.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/TokenRhythm/opensquilla --skill paper-refbib-stub命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Query scholarly metadata and the web through Crossref, Brave, Tavily, and DuckDuckGo with a single CLI surface. Trigger when the user asks for research search, fact lookup, source discovery, or engine comparison. Results retain DOI, publication year, and authors when supplied by Crossref, then deduplicate by DOI, arXiv ID, or normalized URL. API-key engines gate themselves on the relevant environment variable; Crossref and DuckDuckGo need no key.
Deterministic OpenRouter image generation adapter for Nano Banana Pro / Gemini image models. Use as skill_exec when a meta-skill needs local image files and structured IMAGE_READY records without spawning an LLM agent.
Generate or edit a single image via OpenRouter (google/gemini-3.1-flash-image-preview by default). Accepts a text prompt and optional --input-image for image-to-image editing. Trigger when the user asks for an AI image, illustration, concept art, product render, or wants to modify an existing image.
基于 SOC 职业分类
正在显示 SKILL.md
| name | paper-refbib-stub |
| description | Convert normalized multi-search-engine results to minimal BibTeX, preserving real DOI, author, publication year, arXiv ID, source URL, and provenance metadata when available. Unknown years are omitted and duplicate DOI records collapse to one entry. |
| description_zh | 将规范化的 multi-search-engine 结果转换为最小 BibTeX；在可用时保留真实 DOI、作者、出版年份、arXiv ID、来源 URL 和溯源元数据。未知年份不写入，重复 DOI 合并为一条。 |
| user-invocable | false |
| disable-model-invocation | true |
| provenance | {"origin":"opensquilla-original","license":"Apache-2.0"} |
| metadata | {"platform":{"emoji":"📚","requires":{"anyBins":"[Truncated]"}}} |
| entrypoint | {"command":"python {baseDir}/scripts/json_to_bib.py","args":["--out","paper/references.bib"],"stdin":"{{ outputs.search_papers }}","parse":"text","timeout":10} |
Reads a multi-search-engine JSON document on stdin and emits a BibTeX file
of @misc{} entries keyed ref1, ref2, ... Caller wires the upstream
search output via entrypoint.stdin.
The converter consumes optional normalized doi, authors,
corporate_authors, and year fields, while retaining URL-based DOI/arXiv
detection for older producers. Corporate authors are protected with nested
BibTeX braces so institution names are not rearranged as personal names.
It never invents a publication year: missing or malformed years are omitted.
DOIs are normalized case-insensitively, and repeated DOI records emit only
the first entry so citation keys stay unique and deterministic.
All search-provided text fields are treated as untrusted plain text. Control characters, backticks, embedded BibTeX entry fragments, backslashes, and unbalanced braces are neutralized before field emission; structural URL characters are percent-encoded without discarding Unicode locators. This keeps truncated search snippets from corrupting the generated BibTeX database.
