lamm-mit/scholar-search

Google Scholar Search

Search Google Scholar via SerpAPI for academic papers, conference proceedings, and theses. Returns titles, authors, venues, citation counts, and links. Useful for finding recent research on critical minerals, supply chains, extraction methods, and materials science.

Usage

Basic search:

python3 {baseDir}/scripts/scholar_search.py --query "lithium extraction brine"

Filter by year range:

python3 {baseDir}/scripts/scholar_search.py --query "rare earth separation" --year-from 2020 --year-to 2025

Sort by citations:

python3 {baseDir}/scripts/scholar_search.py --query "cobalt supply chain" --sort-by citations

JSON output:

python3 {baseDir}/scripts/scholar_search.py --query "critical minerals policy" --format json --num-results 15

Parameters

| Parameter | Description | Default | |-----------|-------------|---------| | --query | Search query | Required | | --year-from | Start year for publication filter | - | | --year-to | End year for publication filter | - | | --num-results | Number of results to return (max 20) | 10 | | --sort-by | Sort order: relevance, citations | relevance | | --format | Output format: summary, detailed, json | summary |

Examples

# Recent papers on lithium recycling
python3 {baseDir}/scripts/scholar_search.py --query "lithium-ion battery recycling" --year-from 2022

# Most cited papers on rare earth processing
python3 {baseDir}/scripts/scholar_search.py --query "rare earth element processing" --sort-by citations --num-results 15

# Critical minerals policy research
python3 {baseDir}/scripts/scholar_search.py --query "critical minerals supply chain policy" --format detailed

# Cobalt extraction techniques in JSON
python3 {baseDir}/scripts/scholar_search.py --query "cobalt extraction hydrometallurgy" --format json

Notes

• Requires SERPAPI_KEY environment variable
• Google Scholar results include citation counts for impact assessment
• Maximum 20 results per query (SerpAPI limitation)
• For DOE-specific research, use the osti-database skill instead
• For local corpus search, use the corpus-search skill