Librarium -- Multi-Provider Deep Research

Coordinate multi-provider deep-research queries and write markdown outputs.

128|4|Updated Feb 20, 2026

Search Tags:#citations #multi-provider #deep-research #parallel-processing #librarium

Ready to run

npx skills add https://github.com/jkudish/librarium --skill librarium-multi-provider-deep-research

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: Librarium -- Multi-Provider Deep Research
Source: https://github.com/jkudish/librarium/tree/main
Command: npx skills add https://github.com/jkudish/librarium --skill librarium-multi-provider-deep-research

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Librarium enables researchers to execute multi-provider deep-research queries in parallel, surfacing diverse sources and consolidated insights.

Core Features & Use Cases

Parallel execution of deep-research providers to accelerate discovery
Deduplication of citations and structured per-provider outputs
Use Case: Conduct a thorough literature review by synthesizing findings from multiple sources and citations

Quick Start

Ask Librarium to run a deep-research query across configured providers and save the results to the output directory.

Frequently Asked Questions about Librarium -- Multi-Provider Deep Research

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I run deep-research queries across multiple providers in parallel?▼

You can execute multi-provider deep-research in parallel by applying a defined set of providers deterministically, with options to group, wait for completion, or retrieve results as needed.

What is the best way to deduplicate citations from multiple research sources?▼

Deduplicating citations from diverse sources is handled automatically during deterministic orchestration, which also writes per-provider markdown outputs and a run.json manifest to track findings.

How does parallel processing work for deep-research literature reviews?▼

Parallel processing for deep-research works by coordinating queries across diverse providers simultaneously, accelerating discovery and surfacing a consolidated summary of synthesized findings.

Can I save per-provider research results to separate markdown files?▼

Yes, per-provider research results are written to separate markdown outputs, alongside a run.json manifest, ensuring structured documentation of findings from each configured source.

Does multi-provider deep-research require any specific dependencies or components?▼

No specific dependencies or components are required to run multi-provider deep-research, as the Skill handles deterministic orchestration and consolidated summary generation natively.

When should I use parallel multi-provider research instead of a single source query?▼

Parallel multi-provider research should be used instead of a single source query when comprehensive findings are needed from diverse sources, such as conducting a thorough literature review with deduplicated citations.