@wentorai/research-plugins 1.0.0

package/skills/research/funding/open-science-guide/SKILL.md

@@ -0,0 +1,255 @@
+---
+name: open-science-guide
+description: "Pre-registration, open data, and FAIR principles for research"
+metadata:
+  openclaw:
+    emoji: "unlock"
+    category: "research"
+    subcategory: "funding"
+    keywords: ["pre-registration", "registered report", "open science framework", "FAIR data principles", "open data"]
+    source: "wentor-research-plugins"
+---
+
+# Open Science Guide
+
+Implement open science practices including study pre-registration, open data sharing, registered reports, and FAIR data principles to increase research transparency and reproducibility.
+
+## Why Open Science?
+
+Open science practices address the replication crisis and increase trust in research findings:
+
+| Practice | Problem It Addresses |
+|----------|---------------------|
+| Pre-registration | Prevents HARKing (hypothesizing after results are known) and p-hacking |
+| Open data | Enables verification, reanalysis, and meta-analyses |
+| Open materials | Allows exact replication of studies |
+| Open access | Removes paywalls that limit access to knowledge |
+| Registered reports | Eliminates publication bias (acceptance before results are known) |
+| Open code | Enables computational reproducibility |
+
+## Pre-Registration
+
+### What to Pre-Register
+
+Pre-registration commits you to your research plan before seeing the data:
+
+```markdown
+Pre-registration template (standard fields):
+
+1. HYPOTHESES
+   - H1: [Specific, directional hypothesis]
+   - H2: [Another hypothesis]
+
+2. DESIGN
+   - Study type: [Experiment / Survey / Observational]
+   - Between/within subjects design: [Details]
+   - Conditions: [List experimental conditions]
+
+3. SAMPLING PLAN
+   - Sample size: [N = X, justified by power analysis]
+   - Stopping rule: [When will data collection stop?]
+   - Inclusion/exclusion criteria: [List]
+
+4. VARIABLES
+   - Independent variables: [List with levels]
+   - Dependent variables: [List with measurement details]
+   - Covariates: [List any control variables]
+
+5. ANALYSIS PLAN
+   - Primary analysis: [Exact statistical test, e.g., "2x3 mixed ANOVA"]
+   - Secondary analyses: [Additional planned analyses]
+   - Inference criteria: [alpha level, correction for multiple comparisons]
+   - Exclusion criteria: [How will outliers or failed attention checks be handled?]
+   - Missing data: [How will missing data be handled?]
+
+6. OTHER
+   - Exploratory analyses: [Analyses not tied to specific hypotheses]
+```
+
+### Where to Pre-Register
+
+| Platform | URL | Disciplines | Features |
+|----------|-----|-------------|----------|
+| OSF Registries | osf.io/registries | All | Free, flexible templates, versioned |
+| AsPredicted | aspredicted.org | Social sciences, psychology | Simple 9-question form, private until shared |
+| ClinicalTrials.gov | clinicaltrials.gov | Clinical research | Required for clinical trials (FDA) |
+| PROSPERO | crd.york.ac.uk/prospero | Systematic reviews | Health-related reviews only |
+| AEA RCT Registry | socialscienceregistry.org | Economics | RCTs in social sciences |
+
+### Pre-Registration Workflow
+
+```
+1. Design your study
+2. Write the pre-registration document
+3. Have a colleague review it
+4. Submit to a registration platform
+5. Receive a time-stamped registration (URL + DOI)
+6. Collect and analyze data following the pre-registered plan
+7. Report results transparently:
+   - Confirmatory analyses (pre-registered)
+   - Exploratory analyses (clearly labeled as exploratory)
+8. Link the pre-registration in your manuscript
+```
+
+## Registered Reports
+
+Registered Reports are a publication format where peer review occurs before data collection:
+
+```
+Stage 1 (Before Data Collection):
+  - Submit introduction, methods, and analysis plan
+  - Peer review evaluates the research question and methodology
+  - If accepted: "In-Principle Acceptance" (IPA)
+  - Paper will be published regardless of results
+
+Stage 2 (After Data Collection):
+  - Collect data following the approved protocol
+  - Analyze and report results
+  - Add discussion section
+  - Final peer review checks adherence to protocol
+  - Publication
+```
+
+Over 300 journals now accept Registered Reports. Check the registry at cos.io/rr.
+
+### Benefits of Registered Reports
+
+- Eliminates publication bias (null results are published)
+- Ensures methodological rigor is reviewed before sunk costs
+- Prevents post-hoc changes to hypotheses or analyses
+- Provides certainty of publication to researchers
+
+## FAIR Data Principles
+
+FAIR principles ensure research data is **F**indable, **A**ccessible, **I**nteroperable, and **R**eusable:
+
+### Findable
+
+```markdown
+- F1: Data are assigned a globally unique, persistent identifier (DOI)
+- F2: Data are described with rich metadata
+- F3: Metadata include the identifier of the data
+- F4: Data are registered or indexed in a searchable resource
+
+Actions:
+- Deposit data in a repository that assigns DOIs
+- Write a comprehensive README and data dictionary
+- Use standard metadata schemas (Dublin Core, DataCite)
+```
+
+### Accessible
+
+```markdown
+- A1: Data are retrievable by their identifier using open protocols (HTTP)
+- A2: Metadata remain accessible even if data are no longer available
+
+Actions:
+- Use established repositories (not personal websites)
+- Specify access conditions clearly (open, restricted, embargoed)
+- Even if data cannot be shared, publish metadata describing them
+```
+
+### Interoperable
+
+```markdown
+- I1: Data use a formal, accessible, shared language (e.g., CSV, JSON, RDF)
+- I2: Data use vocabularies that follow FAIR principles
+- I3: Data include qualified references to other data
+
+Actions:
+- Use standard file formats (CSV, not proprietary Excel)
+- Use standard variable names and coding schemes
+- Link to related datasets using DOIs
+```
+
+### Reusable
+
+```markdown
+- R1: Data are richly described with provenance information
+- R2: Data are released with a clear, accessible data usage license
+- R3: Data meet domain-relevant community standards
+
+Actions:
+- Include a data dictionary with variable descriptions
+- Apply a license (CC-BY 4.0 recommended)
+- Describe data collection procedures, cleaning steps, and known issues
+- Include analysis code alongside data
+```
+
+## Data Sharing Platforms
+
+| Repository | Disciplines | Max Size | DOI | Cost |
+|-----------|-------------|----------|-----|------|
+| Zenodo | All | 50 GB | Yes | Free |
+| Dryad | All (focus on sciences) | Unlimited | Yes | Sliding scale |
+| Figshare | All | 20 GB (free) | Yes | Free/institutional |
+| OSF | All | 5 GB (free) | Yes | Free |
+| Harvard Dataverse | All (focus on social science) | 2.5 GB per file | Yes | Free |
+| ICPSR | Social science | Varies | Yes | Free deposit |
+| GenBank | Genomics | N/A | Accession numbers | Free |
+| Protein Data Bank | Structural biology | N/A | PDB IDs | Free |
+
+## Data Sharing Best Practices
+
+### README Template for Data Deposits
+
+```markdown
+# Dataset: [Title]
+
+## Description
+Brief description of the dataset and the study it comes from.
+
+## Citation
+If you use this data, please cite:
+[Full citation of the associated publication]
+
+## File Description
+- `data_raw.csv` - Raw data as collected (N = 500, 45 variables)
+- `data_processed.csv` - Cleaned data after exclusions (N = 467, 38 variables)
+- `codebook.csv` - Variable descriptions, types, and valid ranges
+- `analysis_script.R` - Complete analysis code reproducing all results
+
+## Variables (data_processed.csv)
+| Variable | Type | Description | Valid Range |
+|----------|------|-------------|-------------|
+| participant_id | string | Unique participant identifier | P001-P500 |
+| age | integer | Age in years | 18-65 |
+| condition | categorical | Experimental condition | control, treatment_a, treatment_b |
+| score_pre | numeric | Pre-test score | 0-100 |
+| score_post | numeric | Post-test score | 0-100 |
+
+## Missing Data
+- 33 participants excluded for failing attention checks
+- 12 missing values in `score_post` (participants did not complete)
+- Missing coded as NA
+
+## License
+CC-BY 4.0 International
+
+## Contact
+[Name, email, ORCID]
+```
+
+### Sensitive Data Considerations
+
+When data cannot be fully shared (e.g., due to participant privacy):
+
+1. **Anonymize**: Remove direct identifiers (name, email, IP) and indirect identifiers (rare combinations of demographics)
+2. **Aggregate**: Share summary statistics or aggregated data instead of individual-level data
+3. **Restricted access**: Deposit data with access controls (e.g., ICPSR restricted-use data)
+4. **Synthetic data**: Generate synthetic datasets that preserve statistical properties
+5. **Controlled access**: Use data use agreements (DUAs) for sensitive data
+6. **Code without data**: At minimum, share analysis code so methods are transparent
+
+## Open Science Badges
+
+Many journals award badges for open science practices:
+
+| Badge | Meaning |
+|-------|---------|
+| Open Data | Data publicly available |
+| Open Materials | Research materials publicly available |
+| Preregistered | Study pre-registered before data collection |
+| Preregistered + Analysis Plan | Preregistered with detailed analysis plan |
+
+These badges (developed by COS) appear on published articles and signal commitment to transparency.

package/skills/research/funding/zenodo-api/SKILL.md

@@ -0,0 +1,174 @@
+---
+name: zenodo-api
+description: "Open research repository for all disciplines"
+metadata:
+  openclaw:
+    emoji: "🔍"
+    category: "research"
+    subcategory: "funding"
+    keywords: ["open data", "data sharing", "data archiving", "FAIR data principles"]
+    source: "https://developers.zenodo.org/"
+    requires:
+      env: ["ZENODO_API_TOKEN"]
+---
+
+# Zenodo API Guide
+
+## Overview
+
+Zenodo is an open-access digital repository developed by CERN and funded by the European Commission. It enables researchers to deposit datasets, software, publications, reports, and any other research-related digital artifacts. Zenodo assigns each upload a DOI, ensuring long-term citability and discoverability of research outputs across all academic disciplines.
+
+The Zenodo API provides programmatic access to the full repository, including the ability to search records, retrieve metadata, create depositions, upload files, and manage communities. It follows FAIR data principles (Findable, Accessible, Interoperable, Reusable) and integrates with GitHub for automatic software archiving.
+
+Researchers, data librarians, and research infrastructure teams use the Zenodo API to automate data publishing workflows, build data discovery tools, integrate archiving into CI/CD pipelines, and harvest metadata for institutional repositories. Zenodo supports versioning, access controls, and rich metadata schemas including DataCite Metadata Schema.
+
+## Authentication
+
+Authentication is required for most write operations and for accessing restricted records. Zenodo uses personal access tokens.
+
+1. Create an account at https://zenodo.org/
+2. Navigate to **Settings** > **Applications** > **Personal access tokens**
+3. Generate a token with appropriate scopes (`deposit:write`, `deposit:actions`)
+4. Include the token in requests via the `access_token` query parameter or `Authorization` header
+
+```bash
+# Query parameter method
+curl "https://zenodo.org/api/records?access_token=YOUR_TOKEN"
+
+# Header method
+curl -H "Authorization: Bearer YOUR_TOKEN" "https://zenodo.org/api/records"
+```
+
+Read-only access to public records does not require authentication, but authenticated requests have higher rate limits.
+
+## Core Endpoints
+
+### records: Search and Retrieve Records
+
+Search the Zenodo repository for published records including datasets, software, publications, and other artifacts.
+
+- **URL**: `GET https://zenodo.org/api/records`
+- **Parameters**:
+
+| Parameter   | Type   | Required | Description                                          |
+|-------------|--------|----------|------------------------------------------------------|
+| q           | string | No       | Search query (Elasticsearch syntax)                  |
+| type        | string | No       | Record type: `dataset`, `software`, `publication`    |
+| communities | string | No       | Filter by community identifier                       |
+| sort        | string | No       | Sort order: `bestmatch`, `mostrecent`                |
+| size        | int    | No       | Number of results per page (default 10, max 9999)    |
+| page        | int    | No       | Page number (default 1)                              |
+
+- **Example**:
+
+```bash
+curl "https://zenodo.org/api/records?q=climate+change&type=dataset&sort=mostrecent&size=5"
+```
+
+- **Response**: Returns `hits` object containing `total` count and array of record objects with `id`, `doi`, `metadata` (title, creators, description, keywords, publication_date), `files` (download links), and `links`.
+
+### depositions: Create and Manage Uploads
+
+Create new depositions to upload research data and software to Zenodo.
+
+- **URL**: `GET https://zenodo.org/api/deposit/depositions`
+- **Parameters**:
+
+| Parameter    | Type   | Required | Description                                     |
+|--------------|--------|----------|-------------------------------------------------|
+| access_token | string | Yes      | Personal access token                           |
+| q            | string | No       | Search within your depositions                  |
+| sort         | string | No       | Sort order: `bestmatch`, `mostrecent`           |
+| size         | int    | No       | Results per page                                |
+
+- **Example**:
+
+```bash
+# List your depositions
+curl "https://zenodo.org/api/deposit/depositions?access_token=YOUR_TOKEN"
+
+# Create a new deposition
+curl -X POST "https://zenodo.org/api/deposit/depositions?access_token=YOUR_TOKEN" \
+  -H "Content-Type: application/json" \
+  -d '{"metadata": {"title": "My Dataset", "upload_type": "dataset", "description": "Sample dataset", "creators": [{"name": "Doe, Jane"}]}}'
+```
+
+- **Response**: Returns deposition object with `id`, `metadata`, `state`, `submitted`, `links` (including bucket URL for file uploads), and `files`.
+
+## Rate Limits
+
+Rate limits vary based on authentication status. Authenticated requests receive higher limits than anonymous requests. Zenodo does not publicly document exact rate limit numbers, but typical guidance is to keep requests under 60 per minute for anonymous access and 100 per minute for authenticated access. The API returns standard HTTP 429 responses when limits are exceeded, with a `Retry-After` header indicating when to retry.
+
+## Common Patterns
+
+### Search for Datasets by Topic
+
+Find openly available datasets in a specific research domain:
+
+```python
+import requests
+
+params = {
+    "q": "machine learning genomics",
+    "type": "dataset",
+    "sort": "mostrecent",
+    "size": 20
+}
+resp = requests.get("https://zenodo.org/api/records", params=params)
+data = resp.json()
+
+for hit in data["hits"]["hits"]:
+    meta = hit["metadata"]
+    print(f"{meta['title']} | DOI: {hit['doi']} | Date: {meta['publication_date']}")
+```
+
+### Archive a GitHub Repository
+
+Zenodo integrates with GitHub to automatically archive releases. You can also deposit programmatically:
+
+```python
+import requests
+
+TOKEN = os.environ["ZENODO_API_TOKEN"]
+headers = {"Authorization": f"Bearer {TOKEN}", "Content-Type": "application/json"}
+
+# Step 1: Create empty deposition
+dep = requests.post("https://zenodo.org/api/deposit/depositions",
+                     headers=headers, json={}).json()
+
+# Step 2: Upload file to the bucket
+bucket_url = dep["links"]["bucket"]
+with open("analysis_code.zip", "rb") as f:
+    requests.put(f"{bucket_url}/analysis_code.zip",
+                 headers={"Authorization": f"Bearer {TOKEN}"},
+                 data=f)
+
+# Step 3: Add metadata and publish
+metadata = {
+    "metadata": {
+        "title": "Analysis Code for Paper XYZ",
+        "upload_type": "software",
+        "description": "Reproducible analysis pipeline.",
+        "creators": [{"name": "Doe, Jane", "affiliation": "University X"}]
+    }
+}
+requests.put(f"https://zenodo.org/api/deposit/depositions/{dep['id']}",
+             headers=headers, json=metadata)
+requests.post(f"https://zenodo.org/api/deposit/depositions/{dep['id']}/actions/publish",
+              headers={"Authorization": f"Bearer {TOKEN}"})
+```
+
+### Harvest Community Records
+
+Retrieve all records from a specific Zenodo community for institutional tracking:
+
+```bash
+curl "https://zenodo.org/api/records?communities=astronomy&size=100&sort=mostrecent"
+```
+
+## References
+
+- Official API documentation: https://developers.zenodo.org/
+- Zenodo homepage: https://zenodo.org/
+- GitHub integration guide: https://guides.github.com/activities/citable-code/
+- DataCite Metadata Schema: https://schema.datacite.org/

package/skills/research/methodology/action-research-guide/SKILL.md

@@ -0,0 +1,201 @@
+---
+name: action-research-guide
+description: "Design and conduct action research and participatory studies"
+metadata:
+  openclaw:
+    emoji: "handshake"
+    category: "research"
+    subcategory: "methodology"
+    keywords: ["action research", "participatory research", "PAR", "community-based research", "practitioner research"]
+    source: "wentor-research-plugins"
+---
+
+# Action Research Guide
+
+A skill for designing and conducting action research (AR), participatory action research (PAR), and community-based participatory research (CBPR). Covers the cyclical AR process, stakeholder engagement, data collection within action cycles, and balancing rigor with practical impact.
+
+## What Is Action Research?
+
+### Defining Characteristics
+
+```
+Action research is research conducted WITH and FOR participants,
+not ON them. It aims to produce both practical improvements and
+new knowledge simultaneously.
+
+Key principles:
+  1. Collaboration: Researchers and participants work as partners
+  2. Cyclical process: Plan -> Act -> Observe -> Reflect -> Repeat
+  3. Practical focus: Solving a real problem in a real setting
+  4. Democratic: Participants have voice in research design
+  5. Reflexive: Ongoing critical reflection on the process
+
+Contrast with traditional research:
+  Traditional: Researcher studies participants at arm's length
+  Action research: Researcher and participants co-investigate
+```
+
+### Types of Action Research
+
+| Type | Focus | Who Leads |
+|------|-------|----------|
+| Practical AR | Improving professional practice | Practitioner-researcher |
+| Participatory AR (PAR) | Empowerment and social change | Community members + researcher |
+| CBPR | Health equity and community needs | Community-academic partnership |
+| Educational AR | Improving teaching and learning | Teachers |
+| Organizational AR | Improving organizational processes | Internal change agents |
+
+## The Action Research Cycle
+
+### Plan-Act-Observe-Reflect
+
+```python
+def action_research_cycle(cycle_number: int,
+                          problem: str,
+                          planned_action: str) -> dict:
+    """
+    Structure one cycle of action research.
+
+    Args:
+        cycle_number: Which iteration of the cycle (1, 2, 3...)
+        problem: The problem or question being addressed
+        planned_action: The intervention or change being implemented
+    """
+    cycle = {
+        "cycle": cycle_number,
+        "problem_definition": problem,
+        "phases": {
+            "plan": {
+                "activities": [
+                    "Collaboratively define the problem with stakeholders",
+                    "Review evidence and prior knowledge",
+                    "Design the intervention or action step",
+                    "Determine data collection methods",
+                    "Establish success criteria"
+                ],
+                "planned_action": planned_action
+            },
+            "act": {
+                "activities": [
+                    "Implement the planned action",
+                    "Document the implementation process",
+                    "Note deviations from the plan and why they occurred",
+                    "Maintain a reflective journal"
+                ]
+            },
+            "observe": {
+                "activities": [
+                    "Collect data during and after the action",
+                    "Gather participant feedback",
+                    "Record outcomes (intended and unintended)",
+                    "Compile evidence of change or no change"
+                ],
+                "data_sources": [
+                    "Observations and field notes",
+                    "Interviews with participants",
+                    "Surveys or questionnaires",
+                    "Artifacts (documents, records, student work)",
+                    "Quantitative measures (if applicable)"
+                ]
+            },
+            "reflect": {
+                "activities": [
+                    "Analyze collected data",
+                    "Discuss findings with stakeholders",
+                    "Identify what worked and what did not",
+                    "Determine modifications for the next cycle",
+                    "Document lessons learned"
+                ]
+            }
+        },
+        "next_cycle": (
+            "Revise the plan based on reflections and begin the next cycle. "
+            "Each cycle should show progressive refinement of the intervention."
+        )
+    }
+
+    return cycle
+```
+
+## Stakeholder Engagement
+
+### Building Genuine Partnerships
+
+```
+Spectrum of Participation (from low to high):
+
+  Inform:       One-way communication to participants
+  Consult:      Gather input, researcher makes decisions
+  Involve:      Participants contribute to some decisions
+  Collaborate:  Shared decision-making throughout
+  Empower:      Participants lead, researcher supports
+
+True action research operates at the Collaborate or Empower level.
+
+Strategies for genuine engagement:
+  - Hold meetings at times and locations convenient for participants
+  - Use accessible language (avoid academic jargon)
+  - Share data and findings openly with all partners
+  - Negotiate authorship and credit transparently
+  - Build relationships before starting the research
+  - Compensate participants for their time and expertise
+```
+
+## Data Collection in Action Research
+
+### Mixed Methods Within Cycles
+
+```
+Qualitative data:
+  - Reflective journals (researcher and participants)
+  - Focus groups after each action cycle
+  - Semi-structured interviews with key stakeholders
+  - Photographs and video documentation
+  - Meeting minutes and decision logs
+
+Quantitative data:
+  - Pre/post surveys measuring target outcomes
+  - Performance metrics (test scores, health indicators)
+  - Process metrics (attendance, participation rates)
+  - Time series data across multiple cycles
+
+Integration:
+  - Use quantitative data to measure change
+  - Use qualitative data to understand WHY change did or did not occur
+  - Triangulate across sources for credibility
+```
+
+## Ensuring Rigor
+
+### Quality Criteria for Action Research
+
+| Criterion | Strategy |
+|-----------|---------|
+| Outcome validity | Did the action solve the problem? |
+| Process validity | Were methods adequate and appropriate? |
+| Democratic validity | Were all stakeholders' perspectives included? |
+| Catalytic validity | Did the research energize participants to act? |
+| Dialogic validity | Has the work been peer-reviewed or scrutinized? |
+
+### Common Critiques and Responses
+
+```
+Critique: "It is not generalizable."
+Response: Action research produces transferable insights, not
+          statistical generalizations. Thick description allows
+          readers to judge applicability to their own context.
+
+Critique: "The researcher is biased (they are part of the setting)."
+Response: AR is transparent about the researcher's positionality.
+          Reflexive journaling, peer debriefing, and member checking
+          mitigate bias. Insider knowledge is a strength, not a flaw.
+
+Critique: "It is just practice improvement, not real research."
+Response: AR produces systematic, evidence-based knowledge that is
+          publicly shared and peer-reviewed. The dual focus on action
+          and research is by design, not a limitation.
+```
+
+## Reporting Action Research
+
+Follow the ARCS (Action Research Cycle Reporting Standards) or use the general structure: describe the context and problem, explain the partnership and roles, present each cycle (plan-act-observe-reflect) chronologically, show how the intervention evolved across cycles, report both practical outcomes and theoretical contributions, and reflect on limitations and what you would do differently.