Castorini
Official@castorini · Waterloo, Ontario, Canada
Jimmy Lin's research group at the University of Waterloo
Agent Skills by Castorini
Showing 6 vetted skills indexed across 1 GitHub repositories.
castorinicastorini-pipeline
Coordinates end-to-end Castorini pipelines across rank_llm, ragnarok, nuggetizer, and umbrela.
castorinianserini-fatjar
Download the latest Anserini fatjar from Maven Central and run CACM smoke tests.
castorinicastorini-cli-reference
Enforce the castorini.cli.v1 CLI contract across Castorini repositories.
castorinicastorini-release
Automate PyPI release workflows for Castorini projects with version bumps and publishing.
castorinicastorini-onboard
Configure Castorini Python repos with uv or pip and run smoke tests.
castorinicastorini-serve
Orchestrate local Anserini retrieval with optional Castorini services via HTTP.
Frequently Asked Questions About Castorini
FAQPage SchemaWhat specific tasks does the Castorini ecosystem enable?▼
Castorini enables end-to-end information retrieval, document ranking, and search indexing. It provides the infrastructure to manage complex retrieval pipelines, execute local search services, and perform standardized benchmarking against datasets like CACM using Lucene-based indexing.
Who is the target persona for these research repositories?▼
The target audience includes information retrieval researchers, data scientists, and academic engineers focused on search system architecture. It is designed for those building, testing, and deploying high-performance ranking models and document retrieval systems within university or research-oriented environments.
What are the prerequisites for running Castorini retrieval services?▼
Users require a Java-based environment to support Anserini fatjar execution and Maven Central dependencies. Additionally, local environments must be configured with package management for repository onboarding and HTTP service orchestration to enable the retrieval and ranking functionality.