workshop

Coordinate parallel researcher agents to interrogate NotebookLM notebooks and synthesize evidence-backed findings.

Updated Feb 28, 2026
One-click install
npx skills add https://github.com/cosmicdreams/claude-plugins --skill workshop-cosmicdreams
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: workshop
Source: https://github.com/cosmicdreams/claude-plugins/tree/main/research-lab/skills/workshop
Command: npx skills add https://github.com/cosmicdreams/claude-plugins --skill workshop-cosmicdreams

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Workshop reduces the time it takes to interrogate a NotebookLM notebook by running multiple facet-focused researcher agents in parallel, then synthesizing their findings into a coherent report.

Core Features & Use Cases

  • Parallel facet research: Splits investigation into 3–5 non-overlapping, answerable facets and assigns each to a dedicated researcher agent.
  • Cross-pollination protocol: Researchers exchange signal-only findings that intersect, contradict, or change priorities, improving overall synthesis quality.
  • Notebook querying workflow: Uses the notebook ask script with 5–8 focused questions per facet and records question/answer/source-level evidence.
  • Output-ready artifacts: Writes per-facet findings to engagement files (03-workshop-N.md) and produces a consolidated synthesis (03-workshop.md).

Quick Start

Give the AI the instruction "Swarm this notebook" along with the Notebook ID and 3–5 research facets, and it will return a synthesized workshop report.

Frequently Asked Questions about workshop

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

FAQPage Schema
How do I run parallel research on a NotebookLM notebook?

Multi-agent parallel interrogation splits your research into 3–5 non-overlapping facets, assigning each to a dedicated researcher agent. This reduces investigation time by querying NotebookLM sources simultaneously rather than sequentially.

How does cross-pollination work in multi-agent research synthesis?

Cross-pollination allows focused researchers to exchange signal-only findings that intersect, contradict, or change priorities. This message exchange protocol improves overall synthesis quality by sharing intersecting evidence across parallel facets.

Can I use multi-agent prompting for solo research investigation?

Yes, you can use multi-agent prompting for a PI-driven solo pass. The system applies the same parallel interrogation, evidence gathering, and cross-pollination protocols to derive findings from existing NotebookLM sources without multiple human researchers.

How many research questions does each focused agent ask per facet?

Each focused researcher agent asks 5–8 focused questions per facet. The notebook querying workflow uses a notebook ask script and records question, answer, and source-level evidence to ensure findings are backed by notebook sources.

What is the best way to synthesize evidence from multiple research facets?

The best way to synthesize evidence from multiple research facets is through multi-agent parallel investigation with a cross-pollination protocol. Dedicated researchers gather evidence per facet, exchange intersecting signals, and consolidate outputs into a single synthesis report.

What are the limitations of multi-agent notebook research?

A limitation of multi-agent notebook research is that facets must be non-overlapping and answerable from existing notebook sources. If the NotebookLM notebook lacks sufficient sources for a specific facet, the researcher cannot derive evidence-backed findings.