bulk_search
Search Google Scholar for multiple queries sequentially with automatic rate limiting.
Use this tool for batch literature reviews when you need results for several topics at once. Each query runs one at a time with randomised delays between them to avoid triggering Google Scholar’s bot detection. Partial failures are captured per-query so a single bad result does not abort the whole batch.
Parameters
Parameter |
Type |
Default |
Description |
|---|---|---|---|
|
|
(required) |
List of search queries. Maximum 10 queries per call. |
|
|
|
Results to return for each query. Clamped to 1–10. |
|
|
|
Filter to papers published in or after this year (applied to all queries). |
|
|
|
Filter to papers published in or before this year (applied to all queries). |
Return Value
Returns a BulkSearchResponse containing:
results— list ofSearchResponse, one per querytotal_queries— number of queries submittedcompleted_queries— number of queries that succeedederrors— list of error strings for any failed queries
Examples
Multi-topic literature sweep:
{
"queries": [
"diffusion models image generation",
"contrastive self-supervised learning",
"neural radiance fields NeRF"
],
"max_results_per_query": 5
}
Year-filtered batch search:
{
"queries": [
"large language model alignment",
"reinforcement learning from human feedback"
],
"max_results_per_query": 8,
"year_from": 2022
}
Warning
Even with rate limiting, large batches (10 queries) can take several minutes.
Prefer smaller batches or lower max_results_per_query when low latency matters.