research-lit
Search and analyze research papers, find related work, summarize key ideas. Use when user says \"find papers\", \"related work\", \"literature review\", \"what does this paper say\", or needs to understand academic papers.
npx skills add majiayu000/claude-skill-registry --skill research-lit-wanshuiyin-auto-claude-code-res --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
The skill conducts a literature review workflow: it searches multiple sources for papers related to a research topic, de-duplicates results across sources, and extracts metadata and summaries. It prioritizes local PDFs first (PAPER_LIBRARY), then external sources (web, arXiv, Semantic Scholar, DeepXiv, Exa) as configured. It includes steps to read local papers (first 3 pages) and build a local knowledge base, then augments findings with online sources and structured metadata when enabled. It supports various source-selection overrides via explicit directives and handles arXiv API results, venue metadata, and deduplication rules.
How it works
- Defines constants for paper library locations, scanning limits (MAX_LOCAL_PAPERS), and ARXIV behavior (ARXIV_DOWNLOAD, ARXIV_MAX_DOWNLOAD).
- Determines sources from a topic directive: if — sources: is present, only search those; if not, default to all, with certain exclusions. Supported sources include zotero, obsidian, local, web, semantic-scholar, deepxiv, exa, all.
- Step 0a (Zotero): If Zotero is requested and configured, perform a Zotero search by topic, read collections, extract annotations, export BibTeX, and compile per-entry details (title, authors, year, venue, annotations, tags, collection).
- Step 0b (Obsidian): If Obsidian is requested and configured, search vault, check relevant tags, read summaries, follow wikilinks, and compile per-note details (title, path, summaries, links, metadata).
- Step 0c (Local PDFs): Locate PDFs in PAPER_LIBRARY paths, deduplicate against Zotero results, filter by relevance to topic using filename/first-page content, and for up to MAX_LOCAL_PAPERS, read first 3 pages to extract title, authors, year, core contribution, and relevance; build a local knowledge base.
- Step 1 (External search): Use WebSearch and (optionally) arXiv API, Semantic Scholar, DeepXiv, and Exa according to requested sources. Always attempt arXiv API if ARXIV_FETCHER is available; otherwise fall back to WebSearch. De-duplicate against local and previously found sources.
- arXiv API integration: attempts to resolve an arXiv fetcher script and run a search for the topic to obtain structured results; on failure, degradations are logged but do not halt the workflow.
- Semantic Scholar: optional; if requested, uses a S2 fetcher to obtain venue papers with metadata and citations, and performs deduplication with arXiv results.
- DeepXiv: optional; if requested, uses a DeepXiv fetcher or CLI to retrieve papers progressively, with de-duplication and metadata integration.
- Exa: optional; if requested, uses an Exa fetcher to perform broad AI-powered web search with content extraction and highlights.
- Deduplication rules: prioritize arXiv ID where available, then normalized title; maintain a canonical entry and record additional sources contributing to it.
When to use it
- Use when the user says "find papers", "related work", "literature review", "what does this paper say", or needs to understand academic papers.
- Overrides allow targeting specific sources or behaviors (e.g., local only, web only, arXiv download, etc.).
What it can touch
- Local files: PAPER_LIBRARY paths (papers/, literature/).
- External data sources: Zotero MCP, Obsidian MCP, WebSearch, arXiv API, Semantic Scholar API, DeepXiv CLI/fetcher, Exa fetcher (all controlled by source selection and availability in the environment).
Caveats
- Relies on external tools/configurations; if a requested source is not configured, the skill proceeds with available sources and logs warnings.
- arXiv downloads are controlled by ARXIV_DOWNLOAD and ARXIV_MAX_DOWNLOAD; downloads occur only when explicitly enabled.
- Exa, DeepXiv, and Semantic Scholar integrations are optional and may be unavailable depending on environment or explicit user directives.
# Research Literature Review Research topic: $ARGUMENTS ## Constants - **PAPER_LIBRARY** — Local directory containing user's paper collection (PDFs). Check these paths in order: 1. `papers/` in the current project directory 2. `literature/` in the current project directory 3. Custom path specified by user in `AGENTS.md` under `## Paper Library` - **MAX_LOCAL_PAPERS = 20** — Maximum number of local PDFs to scan (read first 3 pages each). If more are found, prioritize by filename relevance to the topic. - **ARXIV_DOWNLOAD = false** — When `true`, download top 3-5 most relevant arXiv PDFs to PAPER_LIBRARY after search. When `false` (default), only fetch metadata (title, abstract, authors) via arXiv API — no files are downloaded. - **ARXIV_MAX_DOWNLOAD = 5** — Maximum number of PDFs to download when `ARXIV_DOWNLOAD = true`. - **REVIEWER_BACKEND = `codex`** — Default reviewer route for optional literature synthesis cross-checks. Use `--reviewer: oracle-pro` only when explicitly requested; if Oracle is unavailable, warn and continue with Codex xhigh or local synthesis. > 💡 Overrides: > - `/research-lit "topic" — paper library: ~/my_papers/` — custom local PDF path > - `/research-lit "to
- Constants
- Data Sources
- Source Selection
- Source Table
- Workflow
- Step 0a: Search Zotero Library (if available)
- Step 0b: Search Obsidian Vault (if available)
- Step 0c: Scan Local Paper Library
- Step 1: Search (external)
- Step 2: Analyze Each Paper
- Step 3: Synthesize
- Step 4: Output
- Step 5: Save (if requested)
- Step 6: Update Research Wiki
if [ -z "${ARIS_REPO:-}" ] && [ -f .aris/installed-skills-codex.txt ]; then
fi
if [ -n "$ARXIV_FETCHER" ]; then
if python3 "$ARXIV_FETCHER" search "QUERY" --max 10; then
echo "D2 contribution: arxiv (helper invocation exit 0)" >&2
else
echo "WARN: arxiv_fetch.py invocation failed; D2 aggregate continues with WebSearch results." >&2
echo "WARN: arxiv_fetch.py not resolved; falling back to WebSearch for arXiv hits." >&2
Re-resolve $ARIS_REPO + $S2_FETCHER (SKILL bash blocks may run in separate shells).
if [ -n "$S2_FETCHER" ]; thenWhat does the research-lit skill do?
Search and analyze research papers, find related work, summarize key ideas. Use when user says \"find papers\", \"related work\", \"literature review\", \"what does this paper say\", or needs to understand academic papers.
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill research-lit-wanshuiyin-auto-claude-code-res --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.