@wentorai/research-plugins 1.1.0 → 1.2.0

package/skills/research/deep-research/corvus-research-guide/SKILL.md

@@ -0,0 +1,132 @@
+---
+name: corvus-research-guide
+description: "Multi-agent AI research with semantic search and citation snowballing"
+metadata:
+  openclaw:
+    emoji: "🐦‍⬛"
+    category: "research"
+    subcategory: "deep-research"
+    keywords: ["Corvus", "multi-agent", "semantic search", "citation snowballing", "research synthesis", "AI research"]
+    source: "https://github.com/corvus-research/corvus"
+---
+
+# Corvus Research Guide
+
+## Overview
+
+Corvus is a multi-agent AI research system that combines semantic search, forward/backward citation snowballing, and synthesis to conduct thorough literature investigations. It iteratively expands search results by following citation chains, identifies research gaps, and generates structured research briefs with full provenance.
+
+## Architecture
+
+### Agent Pipeline
+
+```
+Query → Semantic Search Agent
+           ↓
+     Citation Snowball Agent (forward + backward)
+           ↓
+     Relevance Filter Agent
+           ↓
+     Synthesis Agent
+           ↓
+     Report with Citation Graph
+```
+
+### Key Features
+
+1. **Semantic search**: Uses embedding-based search across Semantic Scholar, OpenAlex
+2. **Citation snowballing**: Iteratively follows references (backward) and citations (forward) to discover related work
+3. **Relevance scoring**: AI-based relevance assessment at each expansion step
+4. **Provenance tracking**: Every claim linked to source papers
+5. **Gap identification**: Identifies under-explored research areas
+
+## Usage
+
+```python
+from corvus import ResearchAgent
+
+agent = ResearchAgent(
+    llm_provider="anthropic",
+    search_backends=["semantic_scholar", "openalex"],
+    max_snowball_depth=2,
+)
+
+# Conduct deep research
+result = agent.research(
+    query="What are the current approaches to continual learning "
+          "in large language models?",
+    initial_papers=20,
+    snowball_per_paper=5,
+)
+
+# Access results
+print(f"Papers found: {len(result.papers)}")
+print(f"Unique clusters: {len(result.clusters)}")
+print(f"\nSynthesis:\n{result.synthesis}")
+
+# Export citation graph
+result.export_graph("citation_network.gexf")
+
+# Export bibliography
+result.export_bibtex("references.bib")
+```
+
+### Snowballing Configuration
+
+```python
+agent = ResearchAgent(
+    snowball_config={
+        "max_depth": 3,           # Citation chain depth
+        "backward_limit": 10,     # References per paper
+        "forward_limit": 10,      # Citations per paper
+        "relevance_threshold": 0.7,  # Min relevance to continue
+        "year_filter": 2020,      # Only papers from 2020+
+    }
+)
+```
+
+### Research Modes
+
+```python
+# Broad survey mode
+result = agent.research(query, mode="survey",
+                        initial_papers=50)
+
+# Focused deep-dive
+result = agent.research(query, mode="focused",
+                        initial_papers=10,
+                        snowball_depth=3)
+
+# Gap analysis
+result = agent.research(query, mode="gap_analysis")
+# Returns underexplored subtopics and suggested directions
+```
+
+## Output Format
+
+```python
+# Structured research brief
+brief = result.generate_brief()
+
+# Contains:
+# - Research question
+# - Methodology (search strategy, databases, snowball depth)
+# - Key themes (clustered by topic)
+# - Timeline (research evolution over time)
+# - Gap analysis (underexplored areas)
+# - Bibliography (all papers with citation counts)
+
+brief.save("research_brief.md")
+```
+
+## Use Cases
+
+1. **Literature reviews**: Comprehensive coverage via snowballing
+2. **Research gap identification**: Find underexplored subtopics
+3. **Trend analysis**: Track research evolution through citation chains
+4. **Grant proposals**: Quick evidence of research need
+
+## References
+
+- [Corvus GitHub](https://github.com/corvus-research/corvus)
+- Wohlin, C. (2014). "Guidelines for snowballing in systematic literature studies." *EASE 2014*.

package/skills/research/deep-research/in-depth-research-guide/SKILL.md

@@ -0,0 +1,205 @@
+---
+name: in-depth-research-guide
+description: "Structured methodology for conducting exhaustive multi-source investigations"
+metadata:
+  openclaw:
+    emoji: "🔬"
+    category: "research"
+    subcategory: "deep-research"
+    keywords: ["deep research", "systematic investigation", "multi-source research", "evidence synthesis", "research methodology", "source evaluation"]
+    source: "https://clawhub.ai/ivangdavila/in-depth-research"
+---
+
+# In-Depth Research Methodology
+
+## Overview
+
+In-depth research goes beyond surface-level literature review to conduct exhaustive, multi-source investigations that synthesize evidence from academic papers, grey literature, industry reports, datasets, and primary sources. This methodology is used when a research question requires comprehensive coverage — for systematic reviews, policy briefs, competitive analyses, or foundational literature surveys in a new research direction.
+
+## The 5-Phase Investigation Framework
+
+### Phase 1: Scope Definition (10% of effort)
+
+Before searching, define boundaries explicitly:
+
+```markdown
+## Research Brief Template
+
+**Central Question**: [One sentence, specific and falsifiable]
+**Sub-Questions** (3-5):
+  1. [Decomposed aspect 1]
+  2. [Decomposed aspect 2]
+  3. [Decomposed aspect 3]
+
+**Inclusion Criteria**:
+  - Time range: [e.g., 2018-present]
+  - Languages: [e.g., English, Chinese]
+  - Document types: [peer-reviewed, preprints, reports, patents]
+  - Disciplines: [e.g., CS, cognitive science, linguistics]
+
+**Exclusion Criteria**:
+  - [Opinion pieces, blog posts without data]
+  - [Studies with n < 30 unless qualitative]
+  - [Duplicate publications of same study]
+
+**Expected Deliverable**: [Literature review / Evidence map / Policy brief / State-of-art report]
+**Depth Target**: [Exhaustive / Representative / Exploratory]
+```
+
+### Phase 2: Multi-Source Collection (30% of effort)
+
+Search systematically across source tiers:
+
+| Tier | Source Type | Examples | Purpose |
+|------|-----------|---------|---------|
+| **1** | Academic databases | Semantic Scholar, PubMed, Scopus, Web of Science | Peer-reviewed primary research |
+| **2** | Preprint servers | arXiv, bioRxiv, SSRN, medRxiv | Cutting-edge, not yet reviewed |
+| **3** | Grey literature | WHO reports, World Bank, NBER working papers | Policy and institutional knowledge |
+| **4** | Patents and standards | Google Patents, USPTO, IEEE standards | Technical implementations |
+| **5** | Data repositories | Zenodo, Figshare, Kaggle, ICPSR | Raw data and reproducibility |
+| **6** | Expert knowledge | Conference talks, interviews, personal communication | Tacit knowledge, emerging trends |
+
+**Search strategy per source**:
+
+```markdown
+For each source:
+1. Construct 3-5 query variants (synonyms, related terms, translated terms)
+2. Apply inclusion/exclusion filters
+3. Record: query string, date, results count, relevant hits
+4. Download and tag all relevant items
+5. Snowball: check references of key papers (backward) and citing papers (forward)
+```
+
+### Phase 3: Source Evaluation (20% of effort)
+
+Rate each source on a standardized evidence hierarchy:
+
+```
+Level 1: Systematic reviews and meta-analyses
+Level 2: Randomized controlled trials / controlled experiments
+Level 3: Cohort studies / quasi-experimental designs
+Level 4: Case-control studies / cross-sectional surveys
+Level 5: Case reports / case series / expert opinion
+Level 6: Anecdotal evidence / grey literature without methodology
+```
+
+**Credibility checklist per source**:
+
+```markdown
+□ Author credentials and affiliation
+□ Publication venue (impact factor, peer-review process)
+□ Methodology transparency (can you replicate it?)
+□ Sample size and representativeness
+□ Conflict of interest disclosure
+□ Recency (is the data still relevant?)
+□ Citation count and reception (supportive vs. critical citations)
+□ Consistency with other sources (does it converge or contradict?)
+```
+
+### Phase 4: Evidence Synthesis (30% of effort)
+
+Organize findings into structured artifacts:
+
+#### Evidence Matrix
+
+| Finding | Source(s) | Evidence Level | Strength | Notes |
+|---------|-----------|---------------|----------|-------|
+| LLMs improve code quality by 20-40% | [A], [B], [C] | Level 2-3 | Strong (convergent) | Effect varies by task complexity |
+| Developers trust AI suggestions less for security-critical code | [D], [E] | Level 4 | Moderate | Small sample sizes |
+| No significant effect on debugging time | [F] | Level 2 | Weak (single study) | Contradicts [A] — needs reconciliation |
+
+#### Contradiction Log
+
+When sources disagree, document systematically:
+
+```markdown
+## Contradiction: Effect of X on Y
+
+**Position A**: X increases Y (Smith 2023, Jones 2024)
+  - Evidence: RCT with n=500, effect size d=0.4
+  - Context: University students, controlled setting
+
+**Position B**: X has no effect on Y (Lee 2024)
+  - Evidence: Field study with n=1200, p=0.34
+  - Context: Industry practitioners, naturalistic setting
+
+**Resolution hypothesis**: The effect is moderated by expertise level.
+  Position A's sample (students) shows the effect;
+  Position B's sample (practitioners) does not.
+  → Need: Study that measures expertise as a moderator.
+```
+
+#### Knowledge Map
+
+Visualize the landscape of your findings:
+
+```
+Central Question
+├── Sub-Q1: [Strong evidence — 8 sources, convergent]
+│   ├── Finding 1.1 (Level 2, 3 sources)
+│   ├── Finding 1.2 (Level 3, 2 sources)
+│   └── Finding 1.3 (Level 4, 3 sources)
+├── Sub-Q2: [Mixed evidence — 5 sources, 1 contradiction]
+│   ├── Finding 2.1 (Level 2, 2 sources)
+│   └── Finding 2.2 ⚠️ CONTRADICTED by Finding 2.3
+├── Sub-Q3: [Weak evidence — 2 sources, emerging area]
+│   └── Finding 3.1 (Level 5, 2 sources)
+└── Unexpected: [Theme that emerged during research]
+    └── Finding 4.1 (Level 3, 1 source) → needs further investigation
+```
+
+### Phase 5: Deliverable Production (10% of effort)
+
+Compile findings into the target deliverable format:
+
+**For a Literature Review**:
+1. Organize by themes (not chronologically)
+2. Synthesize across sources (not paper-by-paper summaries)
+3. Identify gaps explicitly ("No studies have examined...")
+4. State implications for your research
+
+**For a State-of-the-Art Report**:
+1. Current landscape with taxonomy
+2. Key advances and timelines
+3. Open problems and active debates
+4. Future directions with evidence basis
+
+**For a Policy Brief**:
+1. Executive summary (1 paragraph)
+2. Evidence summary (1-2 pages)
+3. Policy options with trade-offs
+4. Recommended action with justification
+
+## Iteration Protocol
+
+Deep research is inherently iterative. After Phase 4, reassess:
+
+```
+After synthesis:
+  □ Are all sub-questions adequately answered?
+  □ Are there new sub-questions that emerged?
+  □ Are there critical gaps requiring additional search?
+  □ Are contradictions resolved or at least documented?
+
+If gaps remain:
+  → Return to Phase 2 with refined queries
+  → Maximum 3 iteration cycles before declaring scope complete
+  → Document what remains unknown (future work)
+```
+
+## Quality Indicators
+
+A well-executed in-depth investigation should demonstrate:
+
+- **Breadth**: Multiple source tiers consulted (not just Google Scholar)
+- **Depth**: Key papers read in full, not just abstracts
+- **Rigor**: Evidence levels assessed, contradictions documented
+- **Transparency**: Search strategy reproducible, decisions justified
+- **Currency**: Most recent relevant work included
+- **Balance**: Competing viewpoints represented fairly
+
+## References
+
+- Petticrew, M., & Roberts, H. (2006). *Systematic Reviews in the Social Sciences*. Blackwell.
+- Grant, M. J., & Booth, A. (2009). "A typology of reviews." *Health Information & Libraries Journal*, 26(2), 91-108.
+- Snyder, H. (2019). "Literature review as a research methodology." *Journal of Business Research*, 104, 333-339.

package/skills/research/deep-research/kosmos-scientist-guide/SKILL.md

@@ -0,0 +1,185 @@
+---
+name: kosmos-scientist-guide
+description: "Claude Code-driven autonomous AI Scientist for discovery"
+metadata:
+  openclaw:
+    emoji: "🔭"
+    category: "research"
+    subcategory: "deep-research"
+    keywords: ["AI Scientist", "autonomous discovery", "Claude Code", "research automation", "scientific method", "experiment"]
+    source: "https://github.com/jimmc414/Kosmos"
+---
+
+# Kosmos AI Scientist Guide
+
+## Overview
+
+Kosmos is a Claude Code-driven AI Scientist framework that automates the scientific discovery process — from hypothesis generation through literature review, experiment design, code implementation, result analysis, and paper writing. It uses Claude Code as the execution engine with structured prompts that guide it through the full scientific method. Designed for ML/AI researchers automating experiment pipelines.
+
+## Scientific Pipeline
+
+```
+Research Question
+      ↓
+  Literature Review (search + synthesize)
+      ↓
+  Hypothesis Generation (testable predictions)
+      ↓
+  Experiment Design (variables, controls, metrics)
+      ↓
+  Implementation (code, data pipeline)
+      ↓
+  Execution (run experiments)
+      ↓
+  Analysis (statistics, visualization)
+      ↓
+  Interpretation (findings, limitations)
+      ↓
+  Paper Draft (LaTeX manuscript)
+```
+
+## Project Configuration
+
+```markdown
+# CLAUDE.md for Kosmos AI Scientist
+
+## Research Protocol
+You are an AI Scientist conducting rigorous research.
+Follow the scientific method strictly:
+
+1. **Literature Review**: Search for related work before
+   proposing anything new. Use Semantic Scholar API.
+2. **Hypothesis**: State falsifiable hypotheses clearly.
+3. **Experiment Design**: Define independent/dependent
+   variables, controls, evaluation metrics.
+4. **Implementation**: Write clean, reproducible code.
+   Set random seeds. Log all hyperparameters.
+5. **Analysis**: Run statistical tests. Report confidence
+   intervals, not just point estimates.
+6. **Honesty**: Report negative results. Acknowledge
+   limitations. Never fabricate data.
+
+## Tools Available
+- Python 3.11+ with PyTorch, NumPy, SciPy
+- LaTeX (pdflatex + bibtex)
+- Semantic Scholar API for literature
+- W&B for experiment tracking (optional)
+```
+
+## Workflow Stages
+
+### Stage 1: Literature Review
+
+```python
+# Kosmos automates literature search
+# The AI Scientist searches, reads, and synthesizes
+
+# Guided prompt pattern:
+"""
+Search for papers on: [TOPIC]
+
+1. Find 20+ relevant papers from last 3 years
+2. Read abstracts and identify key methods
+3. Create a summary table:
+   | Paper | Method | Dataset | Key Result |
+4. Identify gaps in current research
+5. Propose novel directions based on gaps
+"""
+```
+
+### Stage 2: Experiment Design
+
+```python
+# Structured experiment specification
+experiment_spec = {
+    "hypothesis": "Sparse attention patterns learned via "
+                  "Gumbel-Softmax outperform fixed patterns "
+                  "on long-sequence tasks",
+    "independent_vars": ["attention_pattern_type"],
+    "dependent_vars": ["accuracy", "throughput", "memory"],
+    "controls": {
+        "model_size": "same parameter count",
+        "training_data": "same dataset and splits",
+        "hyperparams": "same learning rate schedule",
+    },
+    "datasets": ["Long Range Arena", "PG-19"],
+    "baselines": ["full_attention", "local_window",
+                   "linformer", "performer"],
+    "metrics": {
+        "primary": "accuracy",
+        "secondary": ["wall_clock_time", "peak_memory"],
+    },
+    "statistical_tests": ["paired_t_test", "bootstrap_ci"],
+    "seed_runs": 5,
+}
+```
+
+### Stage 3: Implementation and Execution
+
+```python
+# The AI Scientist writes and runs experiment code
+
+# Pattern: iterative implementation with testing
+"""
+Implement the experiment:
+1. Write model code with unit tests
+2. Write training loop with logging
+3. Run small-scale validation (1 epoch, subset)
+4. Verify metrics are computed correctly
+5. Run full experiments (all seeds, all baselines)
+6. Save results to results/ directory
+"""
+
+# Results structure
+# results/
+# ├── config.json         # Full hyperparameters
+# ├── metrics.csv         # All run metrics
+# ├── figures/            # Generated plots
+# └── checkpoints/        # Model checkpoints
+```
+
+### Stage 4: Analysis and Paper
+
+```python
+# Automated analysis and writing
+"""
+Analyze results and write paper:
+1. Compute mean ± std across seeds
+2. Run statistical significance tests
+3. Generate publication-quality figures
+4. Write LaTeX paper with:
+   - Introduction (motivation + contributions)
+   - Related Work (from literature review)
+   - Method (formal description)
+   - Experiments (setup + results + analysis)
+   - Conclusion (summary + limitations + future)
+5. Verify all citations are real (Semantic Scholar)
+"""
+```
+
+## Safety and Ethics
+
+```markdown
+### Guardrails
+- Never fabricate or manipulate experimental data
+- Report all results including negative ones
+- Acknowledge limitations explicitly
+- Verify all citations against real databases
+- Include compute cost and environmental impact
+- Flag when results are inconclusive
+- Human review required before submission
+```
+
+## Use Cases
+
+1. **ML experiments**: Automated hypothesis → experiment → paper
+2. **Ablation studies**: Systematic component analysis
+3. **Baseline comparison**: Reproduce and compare methods
+4. **Research acceleration**: Draft experiments faster
+5. **Teaching**: Demonstrate scientific method with AI
+
+## References
+
+- [Kosmos GitHub](https://github.com/jimmc414/Kosmos)
+- [The AI Scientist](https://arxiv.org/abs/2408.06292)
+- [Claude Code](https://docs.anthropic.com/en/docs/claude-code)

package/skills/research/deep-research/llm-scientific-discovery-guide/SKILL.md

@@ -0,0 +1,178 @@
+---
+name: llm-scientific-discovery-guide
+description: "Survey of LLM agents for biomedical scientific discovery"
+metadata:
+  openclaw:
+    emoji: "🧬"
+    category: "research"
+    subcategory: "deep-research"
+    keywords: ["LLM agents", "scientific discovery", "biomedical AI", "drug discovery", "hypothesis generation", "lab automation"]
+    source: "https://github.com/zjlrock777/Awesome-LLM-Agents-Scientific-Discovery"
+---
+
+# LLM Agents for Scientific Discovery Guide
+
+## Overview
+
+A curated survey of how LLM-based agents are being applied to scientific discovery, with a focus on biomedical research. Covers hypothesis generation, experiment design, lab automation, literature synthesis, and multi-agent scientific collaboration. Tracks papers, tools, and frameworks across the spectrum from fully autonomous to human-in-the-loop systems.
+
+## Landscape
+
+```
+LLM Agents for Scientific Discovery
+├── Hypothesis Generation
+│   ├── Literature-based (gap identification)
+│   ├── Data-driven (pattern discovery)
+│   └── Analogy-based (cross-domain transfer)
+├── Experiment Design
+│   ├── Protocol generation
+│   ├── Parameter optimization
+│   └── Control selection
+├── Lab Automation
+│   ├── Robot control (self-driving labs)
+│   ├── Equipment programming
+│   └── Data collection orchestration
+├── Analysis & Interpretation
+│   ├── Statistical analysis
+│   ├── Visualization
+│   └── Result interpretation
+└── Communication
+    ├── Paper writing
+    ├── Presentation generation
+    └── Peer review simulation
+```
+
+## Key Systems
+
+| System | Domain | Capability |
+|--------|--------|-----------|
+| **AI Scientist** | ML/AI | Full paper generation pipeline |
+| **ChemCrow** | Chemistry | Tool-augmented chemical reasoning |
+| **Coscientist** | Chemistry | Autonomous experiment execution |
+| **BioPlanner** | Biology | Experiment protocol generation |
+| **MedAgent** | Medicine | Clinical trial analysis |
+| **GenAgent** | Genomics | Gene expression analysis |
+| **DrugAgent** | Pharma | Drug interaction prediction |
+
+## Hypothesis Generation
+
+```python
+# LLM-based hypothesis generation pattern
+from scientific_agent import HypothesisGenerator
+
+generator = HypothesisGenerator(
+    llm_provider="anthropic",
+    knowledge_sources=["pubmed", "semantic_scholar"],
+)
+
+hypotheses = generator.generate(
+    domain="oncology",
+    context="Recent findings show that gut microbiome "
+            "composition correlates with immunotherapy response",
+    constraints=[
+        "Must be testable in vitro",
+        "Should involve specific bacterial species",
+        "Must have measurable endpoints",
+    ],
+    num_hypotheses=5,
+)
+
+for h in hypotheses:
+    print(f"\nHypothesis: {h.statement}")
+    print(f"  Rationale: {h.rationale}")
+    print(f"  Supporting evidence: {len(h.evidence)} papers")
+    print(f"  Novelty score: {h.novelty_score:.2f}")
+    print(f"  Feasibility: {h.feasibility}")
+```
+
+## Self-Driving Lab Integration
+
+```python
+# Agent controlling automated experiments
+from scientific_agent import LabAgent
+
+agent = LabAgent(
+    llm_provider="anthropic",
+    equipment=["plate_reader", "liquid_handler", "incubator"],
+    safety_constraints=["bsl2", "max_volume_1ml"],
+)
+
+# Design and run experiment
+result = agent.run_experiment(
+    objective="Determine IC50 of compound X against cell line Y",
+    protocol_type="dose_response",
+    parameters={
+        "compound": "Compound_X",
+        "cell_line": "HeLa",
+        "concentrations": "serial_dilution",
+        "replicates": 3,
+        "readout": "cell_viability",
+    },
+)
+
+print(f"IC50: {result.ic50:.2f} uM")
+print(f"R-squared: {result.r_squared:.3f}")
+result.plot_dose_response("dose_response.pdf")
+```
+
+## Multi-Agent Scientific Collaboration
+
+```python
+# Agents with different scientific roles
+from scientific_agent import ScientificTeam
+
+team = ScientificTeam(
+    agents={
+        "PI": {"role": "research_director",
+               "expertise": "oncology"},
+        "Experimentalist": {"role": "experiment_design",
+                           "expertise": "cell_biology"},
+        "Analyst": {"role": "data_analysis",
+                   "expertise": "biostatistics"},
+        "Writer": {"role": "manuscript_writing",
+                  "expertise": "scientific_communication"},
+    },
+)
+
+# Collaborative research cycle
+project = team.start_project(
+    title="Microbiome-immunotherapy interaction study",
+    timeline_weeks=12,
+)
+
+# Agents collaborate: PI directs → Experimentalist designs →
+# Analyst processes → Writer documents
+```
+
+## Reading Roadmap
+
+```markdown
+### Foundational Papers
+1. "The AI Scientist" (Lu et al., 2024) — Fully automated ML research
+2. "ChemCrow" (Bran et al., 2023) — Chemistry tool-use agent
+3. "Coscientist" (Boiko et al., 2023) — Autonomous chemical research
+4. "BioPlanner" (Biswas et al., 2024) — Biology protocol generation
+
+### Surveys
+5. "Scientific Discovery in the Age of AI" (Wang et al., 2023)
+6. "Foundation Models for Science" (Bommasani et al., 2022)
+7. "LLM Agents: A Survey" (multiple, 2024)
+
+### Ethics & Limitations
+8. "Dual-use concerns of AI in biology" (Sandbrink, 2023)
+9. "Can LLMs Generate Novel Research Ideas?" (Si et al., 2024)
+```
+
+## Use Cases
+
+1. **Literature mining**: Automated hypothesis from research gaps
+2. **Experiment automation**: Self-driving lab orchestration
+3. **Drug discovery**: Multi-agent screening and optimization
+4. **Research planning**: Protocol and proposal generation
+5. **Scientific writing**: Paper drafting with verified claims
+
+## References
+
+- [Awesome-LLM-Agents-Scientific-Discovery](https://github.com/zjlrock777/Awesome-LLM-Agents-Scientific-Discovery)
+- [The AI Scientist](https://arxiv.org/abs/2408.06292)
+- [ChemCrow](https://arxiv.org/abs/2304.05376)