academic-researcher
Academic research assistant for literature reviews, paper analysis, and scholarly writing.
Use this skill for literature collection, screening, and source tracing.
$ npx -y skills add LigphiDonk/Oh-my--paper --skill research-literature-trace --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/research-literature-traceContext preview
The summary Claude sees to decide when to auto-load this skill.
Use this skill for literature collection, screening, and source tracing.
id: research-literature-trace name: Research Literature Trace version: 1.0.0 stages: [survey, ideation] tools: [read_file, search_project, write_file]
Use this skill for literature collection, screening, and source tracing.
1. Prefer official publisher pages, DOI resolvers, conference proceedings, or trusted indexes. 2. Keep claims tied to an explicit source path or URL. 3. Record short screening notes rather than long prose. 4. Mark uncertainty explicitly when a source cannot be verified. 5. Each paper entry in `paper_bank.json` **MUST** include a `url` field pointing to the real paper page (prefer DOI link `https://doi.org/...`; fallback to Semantic Scholar, arXiv, or publisher page). Never leave `url` empty. 6. The `authors` field in `paper_bank.json` **MUST** be an array of strings (e.g. `["Alice", "Bob"]`), never a single string like `"TBD"`. 7. For every markdown artifact (e.g. `domain_map.md`), generate **both** a Chinese version (`domain_map.zh.md`) and an English version (`domain_map.en.md`) alongside the main file.
A Claude Code plugin that turns your terminal into an autonomous research lab — literature survey, experiment execution, paper writing, all in one pipeline.
Academic research assistant for literature reviews, paper analysis, and scholarly writing.
This skill should be used when the user asks to "run initial analysis", "analyze single-cell data", "QC my data", "run…