Parallel Batch Search
SkillSearchProvides utilities for parallelizing batch query search over a pre-built TF-IDF index using the worker initializer pattern to efficiently share the read-only index across worker processes.
Available today. Use it from your connected AI after setup.
No other account needed.
Connect ahel once, and every AI you use reads what you have installed.
Then ask your AI: use the Parallel Batch Search skill
What this skill tells your AI
The instructions your AI receives, as published by openlair/openskill in tasks-evolved/parallel-tfidf-search/environment/skills/evo-parallel-batch-search/SKILL.md and read by ahel’s review.
Overview
Parallelizes batch query search using ProcessPoolExecutor with the worker initializer pattern. The index is passed once per worker at pool creation time (via initargs), avoiding repeated pickling.
Worker Initializer Pattern
init_search_worker(index, documents, top_k)sets module-level globalsworker_search_query(query)reads those globals to perform search- Each query is independent - perfect for data parallelism
Usage
import sys
sys.path.insert(0, '/app/environment/skills/evo-parallel-batch-search/scripts')
from utils import (
init_search_worker,
worker_search_query,
batch_search_parallel
)
# Batch search
results, elapsed = batch_search_parallel(
queries, index, top_k=10, num_workers=4, documents=docs
)
Functions
init_search_worker(index_data, documents_data, top_k)- Worker initializerworker_search_query(query)- Per-query search in worker processbatch_search_parallel(queries, index, top_k, num_workers, documents)- Full pipeline
Signals
- GitHub stars
- 89
- Forks
- 4
- Last commit
- Sep 2026
Advanced
- Catalog kind
- skill
- Gateway key
evo-parallel-batch-search- Source
- github.com/openlair/openskill