---
name: academic-research
description: "Conduct deep academic research for philosophy, neuroscience, cognitive science, and theoretical computer science (computability, complexity, AI theory, logic). Use when user asks to: research academic topics, find scholarly papers, conduct literature reviews, analyze citations, synthesize research findings, explore philosophical arguments, investigate consciousness/cognition, study computability/decidability/Turing machines, or analyze academic debates. Triggers on: 'research papers', 'literature review', 'academic sources', 'scholarly articles', 'philosophy of mind', 'computability theory', 'neuroscience studies', 'find papers on', 'what does the research say'."
---

# Academic Research Skill

Conduct comprehensive academic research mimicking Claude.ai's Research feature, specialized for philosophy, neuroscience, cognitive science, and theoretical CS.

## Research Workflow

### 1. Scope the Query

Before searching, clarify:
- **Domain**: Philosophy / Neuroscience / Cognitive Science / Theoretical CS
- **Depth**: Quick (3-5 sources) | Standard (10-15) | Deep (20+)
- **Focus**: Empirical findings / Theoretical frameworks / Historical development / Current debates

If unclear, ask one clarifying question before proceeding.

### 2. Search Strategy

Use web search with academic-focused queries. Search in waves:

**Wave 1 - Core sources:**
- `"[topic]" site:semanticscholar.org`
- `"[topic]" site:arxiv.org`
- `"[topic]" site:philpapers.org` (for philosophy)
- `"[topic]" site:ncbi.nlm.nih.gov` (for neuroscience)

**Wave 2 - Expand with:**
- `"[topic]" review paper OR survey`
- `"[topic]" [key author name]`
- `"[topic]" [specific journal from references/domains.md]`

**Wave 3 - Follow citations:**
- Search for highly-cited papers found in Wave 1-2
- Look for "cited by" to find recent work building on seminal papers

### 3. Source Evaluation

For each source, extract and assess:
- **Relevance** (0-10): How directly does it address the query?
- **Authority**: Peer-reviewed? Citation count? Author credentials?
- **Recency**: Prioritize last 5 years unless historical context needed
- **Type**: Empirical study / Review / Theoretical / Commentary

Flag preprints (arXiv, bioRxiv) as non-peer-reviewed.

### 4. Triangulation

Cross-reference findings to identify:
- **Consensus**: Claims supported by multiple independent sources
- **Debates**: Conflicting findings or interpretations
- **Gaps**: Underexplored questions
- **Key figures**: Most-cited authors and seminal works

### 5. Synthesis Output

Structure the report as:

```markdown
# Research Report: [Topic]

## Summary
[2-3 paragraph executive summary]

## Key Findings
1. [Finding with citation]
2. [Finding with citation]
...

## Theoretical Landscape
[Major positions, schools of thought, competing frameworks]

## Open Questions
[Active debates, unresolved issues, research gaps]

## Recommended Reading
- [Paper 1] - [1-sentence annotation]
- [Paper 2] - [1-sentence annotation]
...

## References
[Full citations, preferably with DOIs/URLs]
```

## Domain-Specific Guidance

See `references/domains.md` for:
- Key journals and venues per domain
- Important authors and research groups
- Domain-specific terminology
- Relevant arXiv categories

## Citation Format

Default: APA 7th edition. Include:
- DOI when available (as URL: https://doi.org/...)
- arXiv ID for preprints: arXiv:XXXX.XXXXX
- Direct URL to paper when no DOI

## Quality Standards

- Never cite a paper without verifying it exists via search
- Distinguish peer-reviewed from preprints
- Note when findings are contested or preliminary
- Include publication year for temporal context
- Prefer primary sources over secondary summaries

## Subagent Mode

When invoked programmatically, return structured data:

```json
{
  "query": "original research question",
  "domain": "identified domain",
  "sources_found": 15,
  "key_findings": ["finding 1", "finding 2"],
  "consensus_level": "high|moderate|low|contested",
  "top_papers": [
    {"title": "...", "authors": "...", "year": 2023, "url": "..."}
  ],
  "research_gaps": ["gap 1", "gap 2"]
}
```