A production-ready OpenClaw skill library for academic research teams
English · 简体中文 (coming soon)
Note
PaperClaw is a skill library, not a monolithic software package. You can install the full collection or copy only the skill folders that match your team's workflows.
PaperClaw packages 27 production-ready SKILL.md files for academic research team workflows across literature management, synthesis, collaboration, manuscript output, and research tracking. Each skill teaches an OpenClaw-compatible agent when to use a capability, how to invoke it, and what kind of output to produce.
PaperClaw is designed for research teams — labs, collaborative projects, cross-institution groups — who want a shared AI research layer rather than isolated personal tools. It is the knowledge management and collaboration complement to LabClaw's biomedical execution toolkit.
| LabClaw | PaperClaw | |
|---|---|---|
| Focus | Biomedical tools & databases | Cross-field knowledge workflows |
| Unit | Solo researcher | Research team / lab |
| Emphasis | Execution (run analysis) | Synthesis & collaboration |
| Scope | Bio / pharma / med | Any academic field |
| Domain | Skills | Focus |
|---|---|---|
| 📚 Literature | 6 | Living reviews, monitoring, citation graphs, gap detection |
| 🔬 Synthesis | 5 | Contradiction detection, consensus mapping, evidence grading |
| 🤝 Collaboration | 5 | Annotation sharing, expertise mapping, knowledge handoffs |
| ✍️ Output | 6 | Draft generation, grants, rebuttals, presentation decks |
| 📡 Tracking | 5 | Hypothesis versioning, preprint watch, citation alerts |
OpenClaw loads workspace skills from <workspace>/skills. A practical setup:
git clone https://github.com/your-handle/PaperClaw.git
mkdir -p ~/.openclaw/workspace/skills
cp -R PaperClaw/skills/* ~/.openclaw/workspace/skills/Then start a new OpenClaw session so the skill folders are picked up. If you keep your workspace in git, merge only the folders you need rather than copying the full library.
PaperClaw/
├── README.md
└── skills/
├── literature/ # 6 skills: living reviews, arXiv monitoring, citation graphs, gap detection
├── synthesis/ # 5 skills: contradiction detection, consensus mapping, evidence grading
├── collaboration/ # 5 skills: annotation sharing, expertise mapping, knowledge handoffs
├── output/ # 6 skills: draft generation, grants, rebuttals, presentations
└── tracking/ # 5 skills: hypothesis versioning, preprint watch, citation alerts
Tools for managing, discovering, and structuring the academic literature your team reads.
| Skill | Description |
|---|---|
living-review |
Maintains a continuously updated, structured literature review for a research team. Ingests papers from multiple sources and synthesizes findings across the team's collective reading into a living document. |
arxiv-monitor |
Monitors arXiv for new papers matching team research interests. Scores relevance, deduplicates against existing library, and surfaces the most important new work — replacing manual daily browsing. |
citation-graph |
Builds and analyzes citation networks around a body of papers. Reveals foundational works, emerging clusters, and the intellectual lineage of a research area. |
gap-detection |
Analyzes a team's literature base to identify what has NOT been studied, contested, or resolved. Surfaces research gaps, methodological blind spots, and open questions. |
semantic-scholar |
Queries the Semantic Scholar Academic Graph API for paper metadata, citations, author profiles, and semantic search across 200M+ papers. |
zotero-integration |
Connects to a team's Zotero library via the Web API. Reads, writes, and syncs paper collections, annotations, and notes — bridging existing reference workflows with PaperClaw. |
Tools for extracting meaning across papers, not just within them.
| Skill | Description |
|---|---|
contradiction-detection |
Scans a corpus of papers for conflicting empirical claims, methodological disagreements, or opposing conclusions. Surfaces genuine scientific contradictions before the team cites conflicting work. |
consensus-mapping |
Identifies what the field broadly agrees on versus what remains contested. Produces a "state of knowledge" map separating settled science from active debate. |
claim-tracker |
Tracks specific scientific claims across the literature over time — who made the claim, who replicated it, who challenged it, and whether it still stands. |
cross-paper-synthesis |
Synthesizes findings across multiple papers into coherent narrative, structured comparison, or temporal evolution. Produces insights that only emerge from reading across the corpus as a whole. |
evidence-grading |
Evaluates the strength of evidence behind scientific claims based on study design, replication status, venue quality, and recency. Produces structured evidence grades. |
Tools for making a lab's collective knowledge more than the sum of its parts.
| Skill | Description |
|---|---|
annotation-sharing |
Enables a team to share, discuss, and build on PDF annotations across papers. Transforms isolated highlights into a shared knowledge layer. |
expertise-mapping |
Maps expertise distribution across the team based on reading history and annotations. Answers "who knows the most about X?" and identifies knowledge gaps. |
lab-knowledge-handoff |
Packages a departing team member's knowledge into a structured handoff document. Prevents institutional knowledge loss when students graduate or postdocs move on. |
reading-coverage |
Tracks which papers the team has collectively read and what important papers remain unread. Acts as the team's reading dashboard. |
team-sync |
Generates structured lab meeting agendas and paper discussion summaries based on recent team reading activity and open questions. |
Tools for turning research knowledge into manuscripts, grants, and presentations.
| Skill | Description |
|---|---|
draft-generation |
Generates manuscript sections grounded in the team's literature base — with all claims citation-traced and calibrated to evidence grades. |
grant-writing |
Generates agency-specific grant proposal sections (NIH, NSF, ERC, Wellcome) grounded in the team's literature and research history. |
review-response |
Drafts structured, evidence-backed responses to peer reviewer comments. Surfaces relevant literature, recommends strategy, and generates cover letters. |
abstract-writing |
Drafts venue-appropriate abstracts in structured and unstructured formats at multiple word counts, including lay summaries. |
rebuttal-writing |
Drafts concise, strategic rebuttals for conference reviews under tight word limits (NeurIPS, ICML, ICLR, ACL). |
presentation-deck |
Generates structured slide outlines and content for conference talks, lab meetings, and thesis defenses calibrated to duration and audience. |
Tools for maintaining a living record of the team's scientific thinking.
| Skill | Description |
|---|---|
hypothesis-versioning |
Maintains version-controlled registry of team hypotheses — tracking evolution, evidence, and reasoning behind each revision. |
experiment-log |
Maintains a structured, searchable log of experiments linking each run to its hypothesis, motivating papers, and conclusions. |
reproducibility-check |
Evaluates reproducibility of papers or team's own work against venue-specific standards (NeurIPS, Nature, ICML). Flags issues before reviewers do. |
preprint-watch |
Monitors preprint servers for relevant work, tracks preprint-to-publication transitions, and alerts when cited preprints are updated. |
citation-alert |
Monitors citation activity for the team's own papers. Identifies who cites the team's work and classifies citation type (builds on / challenges / compares). |
Every SKILL.md follows a consistent structure:
# Skill Name
## Overview — what this skill enables
## When to Use — trigger conditions for the AI agent
## Key Capabilities — specific tools, APIs, parameters
## Usage Examples — concrete code or workflow examples
## Output Format — what the skill returns
## Notes — caveats, integrations, and tips
| Workflow | Skills involved |
|---|---|
| Start a new research project | arxiv-monitor → living-review → gap-detection → hypothesis-versioning |
| Write a paper | cross-paper-synthesis → draft-generation → abstract-writing → reproducibility-check |
| Respond to reviewers | review-response → evidence-grading → rebuttal-writing |
| Lab meeting prep | reading-coverage → team-sync → annotation-sharing |
| Onboard a new team member | expertise-mapping → reading-coverage → lab-knowledge-handoff |
| Write a grant | gap-detection → consensus-mapping → grant-writing |
| Repository | Why it matters |
|---|---|
wu-yc/LabClaw |
Biomedical execution toolkit — 206 skills for biology, pharma, medicine, and literature search. PaperClaw is the collaboration and synthesis complement. |
openclaw/openclaw |
The main runtime that loads workspace skills and provides the agent workspace model PaperClaw is designed for. |
mims-harvard/ToolUniverse |
Large AI-scientist tool ecosystem. Several PaperClaw skills integrate with ToolUniverse's literature search tools. |
MIT License