kthorn/research-superpower/subagent-driven-review

Main agent:
1. Splits 100 papers into 5 batches of 20
2. Dispatches 5 subagents IN PARALLEL (single message, multiple Task calls)
3. Each subagent:
   - Fetches abstracts for its batch
   - Scores using rubric
   - Returns JSON with results
4. Main agent consolidates results into papers-reviewed.json
 
Time savings: 5x faster than sequential!

I need you to screen papers 1-20 from this PMID list for relevance to [QUERY].
 
PMIDs to screen: [PMID list]
 
Use the evaluating-paper-relevance skill to:
1. Fetch abstract for each PMID
2. Score 0-10 based on:
   - Keywords: [list]
   - Data types needed: [measurements, protocols, datasets, etc.]
3. Return JSON:
 
{
  "screened_papers": [
    {"pmid": "12345", "score": 8, "status": "relevant", "reason": "..."},
    ...
  ],
  "stats": {"highly_relevant": 3, "relevant": 5, "not_relevant": 12}
}
 
Do NOT update papers-reviewed.json - return results only.
 
**Rate limiting (CRITICAL - PubMed limits are SHARED across all parallel subagents):**
- If you are the ONLY subagent running: Use 500ms delays (2 req/sec, safe)
- If running with OTHER parallel subagents: Use longer delays to share capacity
  - You are 1 of 2 parallel: Use 1 second delays
  - You are 1 of 3 parallel: Use 1.5 second delays
  - You are 1 of 5 parallel: Use 2.5 second delays
- If you get HTTP 429 errors: Wait 5 seconds, then use 5-second delays for remaining requests

Main agent:
1. Creates TodoWrite with 15 tasks (one per paper)
2. For each paper, dispatches subagent to:
   - Fetch full text (PMC, Unpaywall)
   - Extract relevant data (tables, figures, methods)
   - Identify key findings
   - Return structured findings
3. Main agent consolidates into SUMMARY.md
4. Reviews and adds to papers-reviewed.json
 
Can dispatch in parallel (5 at a time) or sequentially

Deep dive analysis for paper PMID [12345] / DOI [10.xxxx/yyyy]
 
Use evaluating-paper-relevance skill to:
1. Check for curated data sources (if applicable to domain)
2. Fetch full text (try PMC, then Unpaywall if paywalled)
3. Extract relevant data based on research domain:
   - Data tables and measurements
   - Methods and protocols
   - Key results and findings
   - Figures with relevant information
4. Return structured JSON:
 
{
  "pmid": "12345",
  "doi": "10.xxxx/yyyy",
  "full_text_source": "PMC" or "Unpaywall" or "paywalled",
  "data_sources": ["Table 1", "Figure 3", "Supplementary Data"],
  "key_measurements": ["specific values or ranges found"],
  "methods_summary": "Brief description of methods",
  "key_findings": ["Finding 1", "Finding 2", ...],
  "data_availability": "GEO: GSE12345" or "Code: github.com/..." or null
}
 
Do NOT update papers-reviewed.json - return findings only.

Main agent:
1. Dispatches two subagents IN PARALLEL:
   - Subagent A: Fetch and screen forward citations
   - Subagent B: Fetch and screen backward citations
2. Each returns list of promising PMIDs with scores
3. Main agent:
   - Consolidates results
   - Removes duplicates
   - Adds to screening queue
   - Updates papers-reviewed.json

Find and screen forward citations for PMID [12345].
 
Use traversing-citations skill to:
1. Fetch forward citations from PubMed or OpenCitations
2. Screen abstracts for relevance to [QUERY]
3. Score each citation (0-10)
4. Return JSON with promising papers (score ≥7):
 
{
  "seed_pmid": "12345",
  "direction": "forward",
  "citations_found": 45,
  "relevant_citations": [
    {"pmid": "67890", "score": 8, "title": "...", "reason": "..."},
    ...
  ]
}
 
Do NOT update papers-reviewed.json - return results only.

Subagent extracts:
- GEO/SRA/ENA accessions
- Sample sizes and conditions
- Sequencing methods (RNA-seq, WGS, etc.)
- Analysis pipelines
- Differential expression results

Subagent extracts:
- Algorithm descriptions
- Code repositories (GitHub, GitLab, etc.)
- Benchmark datasets used
- Performance metrics
- Implementation details

Subagent extracts:
- Study design (RCT, cohort, etc.)
- Sample size and demographics
- Intervention details
- Primary outcomes
- Statistical methods

Subagent extracts:
- Study sites and coordinates
- Sampling methods
- Species/taxa studied
- Environmental measurements
- Data repositories

- Screen papers batch 1 (PMIDs 1-20)
- Screen papers batch 2 (PMIDs 21-40)
- Screen papers batch 3 (PMIDs 41-60)
- Screen papers batch 4 (PMIDs 61-80)
- Screen papers batch 5 (PMIDs 81-100)
- Consolidate all subagent results
- Generate SUMMARY.md from consolidated data

I'm dispatching 5 subagents in parallel to screen 100 papers.
 
[Uses Task tool 5 times in single message]

{
  "10.1234/example.2023": {
    "pmid": "12345",
    "status": "highly_relevant",
    "score": 9,
    "source": "pubmed_search_batch1",
    "screened_by": "subagent",
    "timestamp": "2025-10-11T14:30:00Z",
    "found_data": ["measurements", "methods", "datasets"]
  }
}

For each highly relevant paper:
  - Dispatch one subagent per paper
  - Subagent does deep dive extraction
  - Main agent consolidates findings immediately
  - Updates SUMMARY.md progressively

Batch 1: Papers 1-5 (dispatch 5 subagents in parallel)
Wait for completion, consolidate
Batch 2: Papers 6-10 (dispatch 5 subagents in parallel)
Wait for completion, consolidate
...

# Pseudo-code for consolidation
all_results = []
for subagent_output in subagent_results:
    results = parse_json(subagent_output)
    all_results.extend(results['screened_papers'])
# Sort by score
all_results.sort(key=lambda x: x['score'], reverse=True)
# Update papers-reviewed.json
for paper in all_results:
    papers_reviewed[paper['doi']] = {
        'pmid': paper['pmid'],
        'status': paper['status'],
        'score': paper['score'],
        'source': f"subagent_batch_{paper['batch_id']}",
        'timestamp': now()
    }

Dispatch subagent 1 → wait → consolidate → dispatch subagent 2 → wait → consolidate → ...

Dispatch 5 subagents → wait for all 5 → consolidate → dispatch next 5 → ...