@wentorai/research-plugins 1.3.2 → 1.4.2

package/skills/literature/search/paper-search-mcp-guide/SKILL.md

@@ -1,107 +0,0 @@
----
-name: paper-search-mcp-guide
-description: "MCP server for searching papers across arXiv, PubMed, bioRxiv"
-metadata:
-  openclaw:
-    emoji: "🔍"
-    category: "literature"
-    subcategory: "search"
-    keywords: ["paper search", "MCP", "arXiv", "PubMed", "bioRxiv", "multi-database"]
-    source: "wentor-research-plugins"
----
-
-# Paper Search MCP Guide
-
-## Overview
-
-Paper Search MCP is an MCP server that provides unified paper search across multiple academic databases — arXiv, PubMed, bioRxiv, and medRxiv. AI agents can search, download, and analyze papers through a single MCP interface. Handles API authentication, rate limiting, and result normalization across different database formats.
-
-## MCP Configuration
-
-```json
-{
-  "mcpServers": {
-    "paper-search": {
-      "command": "npx",
-      "args": ["@openags/paper-search-mcp"],
-      "env": {}
-    }
-  }
-}
-```
-
-## Available Tools
-
-```markdown
-### search_papers
-Search across all databases or specific ones.
-- query: Search terms
-- source: "arxiv" | "pubmed" | "biorxiv" | "all"
-- max_results: Number of results (default 10)
-- date_from: Start date filter
-- date_to: End date filter
-
-### get_paper
-Get full details for a specific paper.
-- id: arXiv ID, PMID, or DOI
-
-### download_pdf
-Download paper PDF.
-- id: Paper identifier
-- output_dir: Save location
-
-### get_references
-Get references cited by a paper.
-- id: Paper identifier
-
-### get_citations
-Get papers that cite this paper.
-- id: Paper identifier
-```
-
-## Usage in Agent Chat
-
-```markdown
-### Example Interactions
-
-"Search for recent papers on protein folding"
-→ Searches arXiv + PubMed + bioRxiv, returns unified results
-
-"Download the PDF for arxiv:2401.12345"
-→ Fetches and saves PDF locally
-
-"What papers cite the AlphaFold2 paper?"
-→ Looks up forward citations via Semantic Scholar
-
-"Find PubMed papers on CRISPR therapy from 2024"
-→ PubMed-specific search with date filter
-```
-
-## Normalized Result Format
-
-```json
-{
-  "title": "Paper Title",
-  "authors": ["Author A", "Author B"],
-  "abstract": "Abstract text...",
-  "source": "arxiv",
-  "id": "2401.12345",
-  "doi": "10.48550/arXiv.2401.12345",
-  "url": "https://arxiv.org/abs/2401.12345",
-  "published": "2024-01-15",
-  "categories": ["cs.AI", "cs.CL"]
-}
-```
-
-## Use Cases
-
-1. **Literature search**: Multi-database paper discovery
-2. **Agent research**: Automated paper retrieval for AI agents
-3. **PDF collection**: Bulk paper downloading
-4. **Citation analysis**: Forward and backward citation chains
-5. **Cross-database**: Unified search across arXiv + PubMed + bioRxiv
-
-## References
-
-- [paper-search-mcp GitHub](https://github.com/openags/paper-search-mcp)
-- [MCP Specification](https://modelcontextprotocol.io/)

package/skills/research/automation/mcp-server-guide/SKILL.md

@@ -1,211 +0,0 @@
----
-name: mcp-server-guide
-description: "Index of 150 MCP server configs bundled with research-plugins"
-metadata:
-  openclaw:
-    emoji: "🔌"
-    category: "research"
-    subcategory: "automation"
-    keywords: ["MCP", "model-context-protocol", "mcp-server", "tool-integration", "external-services", "research-automation"]
-    source: "wentor-research-plugins"
----
-
-# MCP Server Configurations — Research-Plugins
-
-This plugin bundles 150 ready-to-use MCP (Model Context Protocol) server configurations for academic research workflows. Each config is a JSON file specifying how to install, configure, and connect an external MCP server.
-
-## How to Find MCP Configs
-
-All configs are stored in the `mcp-configs/` directory within the installed plugin:
-
-```
-# Typical installed location
-~/.openclaw/extensions/node_modules/@wentorai/research-plugins/mcp-configs/
-
-# Structure
-mcp-configs/
-├── registry.json          # Machine-readable index of all 150 configs
-├── academic-db/           # 43 configs — academic databases & APIs
-├── reference-mgr/         # 25 configs — citation & reference managers
-├── note-knowledge/        # 23 configs — note-taking & knowledge bases
-├── data-platform/         # 13 configs — data science & experiment tracking
-├── ai-platform/           # 12 configs — LLM orchestration & AI services
-├── dev-platform/          # 12 configs — code hosting & development tools
-├── browser/               #  8 configs — web automation & research agents
-├── communication/         #  4 configs — team messaging (Slack, Discord)
-├── cloud-docs/            #  3 configs — cloud document collaboration
-├── database/              #  3 configs — database connectors (SQLite, Neo4j)
-├── email/                 #  2 configs — email integration (Gmail, IMAP)
-└── repository/            #  2 configs — data repositories (HuggingFace)
-```
-
-## Config File Format
-
-Each JSON file follows this schema:
-
-```json
-{
-  "id": "arxiv-search-mcp",
-  "name": "arXiv Search MCP",
-  "description": "Search and fetch papers from arXiv.org",
-  "category": "reference-mgr",
-  "install": {
-    "runtime": "node",
-    "package": "arxiv-search-mcp",
-    "command": "npm install -g arxiv-search-mcp"
-  },
-  "config": {
-    "env": {}
-  },
-  "tools": ["search", "fetch"],
-  "verified": false,
-  "source": "https://github.com/danimal141/arxiv-search-mcp"
-}
-```
-
-Key fields:
-- **install.command** — shell command to install the MCP server
-- **install.runtime** — `node` or `python`
-- **config.env** — required environment variables (API keys, tokens)
-- **tools** — list of tool names the server exposes
-- **verified** — `true` if tested end-to-end in Research-Claw
-
-## How to Use an MCP Server with OpenClaw
-
-### Step 1: Find the config
-
-```bash
-# Read the registry for a quick overview
-cat ~/.openclaw/extensions/node_modules/@wentorai/research-plugins/mcp-configs/registry.json
-
-# Or browse a specific category
-ls ~/.openclaw/extensions/node_modules/@wentorai/research-plugins/mcp-configs/academic-db/
-```
-
-### Step 2: Install the server package
-
-```bash
-# Node-based server
-npm install -g <package-name>
-
-# Python-based server
-pip install <package-name>
-```
-
-### Step 3: Configure in OpenClaw
-
-Add the MCP server to your `~/.openclaw/openclaw.json`:
-
-```json
-{
-  "mcpServers": {
-    "zotero": {
-      "command": "npx",
-      "args": ["-y", "mcp-zotero"],
-      "env": {
-        "ZOTERO_API_KEY": "your-api-key",
-        "ZOTERO_USER_ID": "your-user-id"
-      }
-    }
-  }
-}
-```
-
-### Step 4: Verify
-
-Restart OpenClaw. The MCP server's tools will appear alongside built-in tools. Test with a simple query to confirm connectivity.
-
-## Category Quick Reference
-
-### Academic Databases (43 configs)
-
-Connectors for scholarly search APIs and paper repositories:
-
-| Config | Description | Runtime |
-|--------|-------------|---------|
-| semantic-scholar-mcp | Semantic Scholar API — papers, authors, citations | node |
-| openalex-mcp | OpenAlex API — works, authors, institutions | node |
-| arxiv-search-mcp | arXiv paper search and retrieval | node |
-| pubmed-mcp | PubMed/MEDLINE biomedical literature | node |
-| crossref-mcp | CrossRef DOI resolution and metadata | node |
-| ... and 38 more | See `mcp-configs/academic-db/` | — |
-
-### Reference Managers (25 configs)
-
-Connect to citation management tools:
-
-| Config | Description | Env Required |
-|--------|-------------|-------------|
-| zotero-mcp | Zotero library search, add, annotate | ZOTERO_API_KEY |
-| mendeley-mcp | Mendeley library access | MENDELEY_TOKEN |
-| google-scholar-mcp | Google Scholar search | — |
-| ... and 22 more | See `mcp-configs/reference-mgr/` | — |
-
-### Note & Knowledge Bases (23 configs)
-
-Connect to knowledge management systems:
-
-| Config | Description | Env Required |
-|--------|-------------|-------------|
-| obsidian-mcp | Obsidian vault read/write | OBSIDIAN_VAULT_PATH |
-| notion-mcp | Notion workspace access | NOTION_API_KEY |
-| logseq-mcp | Logseq graph access | LOGSEQ_GRAPH_PATH |
-| ... and 20 more | See `mcp-configs/note-knowledge/` | — |
-
-### Data & Experiment Platforms (13 configs)
-
-ML experiment tracking and data science tools:
-
-| Config | Description | Env Required |
-|--------|-------------|-------------|
-| wandb-mcp | Weights & Biases experiment tracking | WANDB_API_KEY |
-| jupyter-mcp | Jupyter notebook execution | — |
-| kaggle-mcp | Kaggle datasets and competitions | KAGGLE_KEY |
-| ... and 10 more | See `mcp-configs/data-platform/` | — |
-
-### Other Categories
-
-- **ai-platform/** (12) — OpenAI, HuggingFace, Replicate connectors
-- **dev-platform/** (12) — GitHub, GitLab repository management
-- **browser/** (8) — Playwright, Puppeteer, Firecrawl web automation
-- **communication/** (4) — Slack, Discord, Telegram messaging
-- **cloud-docs/** (3) — Notion, Google Drive, Confluence
-- **database/** (3) — SQLite, PostgreSQL, Neo4j connectors
-- **email/** (2) — Gmail, IMAP/SMTP integration
-- **repository/** (2) — HuggingFace Hub, Dataverse
-
-## Using registry.json Programmatically
-
-The `registry.json` file is a machine-readable index of all configs:
-
-```json
-{
-  "version": "1.1.0",
-  "generated": "2026-03-10",
-  "categories": {
-    "academic-db": { "count": 43, "configs": [...] },
-    "reference-mgr": { "count": 25, "configs": [...] }
-  }
-}
-```
-
-Each entry in `configs[]` has: `id`, `name`, `category`, `file` (relative path to the JSON config).
-
-## Related Skills
-
-- [zotero-mcp-guide](../../writing/citation/zotero-mcp-guide/SKILL.md) — Detailed Zotero MCP setup and workflows
-- [paper-search-mcp-guide](../../literature/search/paper-search-mcp-guide/SKILL.md) — Multi-database paper search MCP
-- [edumcp-guide](../../domains/education/edumcp-guide/SKILL.md) — Education-focused MCP server
-
-## Important Notes
-
-- MCP configs are **reference templates** — you must install the server package separately
-- Some servers require API keys or tokens (check `config.env` in each JSON)
-- Prefer the plugin's built-in Agent Tools (search_papers, resolve_doi, etc.) over MCP for academic database queries — they work out of the box without extra installation
-- The `verified: false` flag means the config has not been end-to-end tested; check the source repository for current status
-
-## References
-
-- [MCP Specification](https://modelcontextprotocol.io)
-- [OpenClaw MCP Documentation](https://docs.openclaw.app)
-- [Registry Source](https://github.com/wentorai/research-plugins/tree/main/mcp-configs)

package/skills/tools/knowledge-graph/paperpile-notion-guide/SKILL.md

@@ -1,84 +0,0 @@
----
-name: paperpile-notion-guide
-description: "Sync Paperpile references and annotations to Notion"
-metadata:
-  openclaw:
-    emoji: "📋"
-    category: "tools"
-    subcategory: "knowledge-graph"
-    keywords: ["Paperpile", "Notion", "sync", "reference management", "annotations", "integration"]
-    source: "https://github.com/gsarti/paperpile-notion"
----
-
-# Paperpile-Notion Integration Guide
-
-## Overview
-
-Paperpile-Notion is a tool that syncs your Paperpile reference library to a Notion database. It exports paper metadata (title, authors, year, DOI, abstract), reading status, labels, and annotations to Notion pages, enabling you to combine reference management with Notion's rich knowledge management features.
-
-## Setup
-
-```bash
-# Install from source
-git clone https://github.com/gsarti/paperpile-notion.git
-cd paperpile-notion
-pip install -e .
-
-# Configuration
-# 1. Export Paperpile library as BibTeX/JSON
-# 2. Create Notion integration and get token
-# 3. Create Notion database with required properties
-# 4. Run sync script
-```
-
-## Notion Database Properties
-
-```markdown
-### Required Properties
-| Property | Type | Source |
-|----------|------|--------|
-| Title | Title | Paper title |
-| Authors | Rich text | Author names |
-| Year | Number | Publication year |
-| DOI | URL | Digital Object Identifier |
-| Abstract | Rich text | Paper abstract |
-| Labels | Multi-select | Paperpile labels/folders |
-| Status | Select | Read / Unread / Reading |
-| Added | Date | Date added to Paperpile |
-| Journal | Rich text | Publication venue |
-| Notes | Rich text | Paperpile notes |
-```
-
-## Sync Workflow
-
-```python
-from paperpile_notion import PaperpileSync
-
-sync = PaperpileSync(
-    paperpile_export="library.bib",    # BibTeX export
-    notion_database_id="your_db_id",
-)
-
-# Full sync
-result = sync.run()
-print(f"Synced: {result.created} new, "
-      f"{result.updated} updated, "
-      f"{result.skipped} unchanged")
-
-# Incremental sync (only new/modified)
-result = sync.run(incremental=True)
-```
-
-## Use Cases
-
-1. **Knowledge management**: Paperpile references in Notion
-2. **Reading tracking**: Kanban board for paper reading status
-3. **Team collaboration**: Shared reference database in Notion
-4. **Research notes**: Rich notes alongside paper metadata
-5. **Project organization**: Tag papers by research project
-
-## References
-
-- [paperpile-notion GitHub](https://github.com/gsarti/paperpile-notion)
-- [Paperpile](https://paperpile.com/)
-- [Notion API](https://developers.notion.com/)

package/src/tools/semantic-scholar.ts

@@ -1,66 +0,0 @@
-import { Type } from "@sinclair/typebox";
-import type { OpenClawPluginApi, OpenClawPluginToolContext } from "openclaw/plugin-sdk";
-import { toolResult } from "./util.js";
-
-const BASE = "https://api.semanticscholar.org/graph/v1";
-
-export function createSemanticScholarTools(
-  ctx: OpenClawPluginToolContext,
-  api: OpenClawPluginApi,
-) {
-  const apiKey =
-    ctx.config?.get?.("plugins.entries.research-plugins.semanticScholarApiKey") ?? "";
-
-  const headers: Record<string, string> = apiKey ? { "x-api-key": apiKey } : {};
-
-  return [
-    {
-      name: "get_paper",
-      label: "Get Paper Details (Semantic Scholar)",
-      description:
-        "Get detailed information about a specific paper by its Semantic Scholar ID, DOI, or ArXiv ID.",
-      parameters: Type.Object({
-        paper_id: Type.String({
-          description:
-            "Paper identifier: Semantic Scholar ID, DOI (e.g. '10.1234/...'), or ArXiv ID (e.g. 'arXiv:2301.00001')",
-        }),
-      }),
-      execute: async (input: { paper_id: string }) => {
-        const fields =
-          "title,abstract,authors,year,citationCount,referenceCount,tldr,url,venue,isOpenAccess,openAccessPdf,fieldsOfStudy,publicationDate";
-        const res = await fetch(
-          `${BASE}/paper/${encodeURIComponent(input.paper_id)}?fields=${fields}`,
-          { headers },
-        );
-        if (!res.ok) return toolResult({ error: `API error: ${res.status} ${res.statusText}` });
-        return toolResult(await res.json());
-      },
-    },
-    {
-      name: "get_citations",
-      label: "Get Citations (Semantic Scholar)",
-      description:
-        "Get papers that cite a given paper. Useful for forward citation tracking.",
-      parameters: Type.Object({
-        paper_id: Type.String({ description: "Paper identifier (S2 ID, DOI, or ArXiv ID)" }),
-        limit: Type.Optional(
-          Type.Number({ description: "Max citations to return (default 20, max 100)" }),
-        ),
-        offset: Type.Optional(
-          Type.Number({ description: "Pagination offset" }),
-        ),
-      }),
-      execute: async (input: { paper_id: string; limit?: number; offset?: number }) => {
-        const fields = "title,authors,year,citationCount,url,abstract";
-        const limit = Math.min(input.limit ?? 20, 100);
-        const offset = input.offset ?? 0;
-        const res = await fetch(
-          `${BASE}/paper/${encodeURIComponent(input.paper_id)}/citations?fields=${fields}&limit=${limit}&offset=${offset}`,
-          { headers },
-        );
-        if (!res.ok) return toolResult({ error: `API error: ${res.status} ${res.statusText}` });
-        return toolResult(await res.json());
-      },
-    },
-  ];
-}