skills.rest
trevorbyrum
trevorbyrumCOMMUNITY
Explore all skills by this author →
GitHub

research-execute

Execute research plans by dispatching parallel subagents across data connectors.

1|1|Updated Mar 9, 2026
Search Tags:#research#AI agents#validation#information retrieval#data collection#synthesis
One-click install
npx skills add https://github.com/trevorbyrum/claude-skills-suite --skill research-execute
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: research-execute
Source: https://github.com/trevorbyrum/claude-skills-suite/tree/main/skills/research-execute
Command: npx skills add https://github.com/trevorbyrum/claude-skills-suite --skill research-execute

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill automates the execution of a research plan by fanning out parallel subagents across various data connectors, synthesizing the findings, and performing a triple-check to identify blind spots and ensure accuracy.

Core Features & Use Cases

  • Parallel Subagent Dispatch: Leverages multiple AI models and tools (Sonnet, Gemini, Codex, Copilot, WebSearch, GitHub, etc.) to gather information concurrently.
  • Data Synthesis and Aggregation: Compiles findings from diverse sources into a coherent summary, including aggregate source counts.
  • Counter-Review: Challenges the synthesized research with different AI models to identify weaknesses, contradictions, or missing information.
  • Use Case: After a research plan is approved, this Skill runs the actual research, gathers data from academic sources, code repositories, and the web, synthesizes the information, and then has other models review it for accuracy and completeness before presenting a final summary.

Quick Start

Use the research-execute skill to run the approved research plan.

Frequently Asked Questions about research-execute

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

FAQPage Schema
How do I execute a research plan using parallel AI agents?▼

To execute a research plan, this Skill dispatches parallel subagents across multiple data connectors to gather information concurrently, synthesizes the findings from diverse sources, and aggregates source counts into a coherent summary.

How does AI counter-review validate research synthesis?▼

Research synthesis is validated through a triple-counter review where different AI models challenge the compiled findings to identify weaknesses, contradictions, or missing information, with feedback integrated into the final summary.

Can I use multiple AI models like Sonnet and Gemini for data collection?▼

Yes, data collection leverages multiple AI models and tools including Sonnet, Gemini, Codex, Copilot, WebSearch, and GitHub to gather information concurrently across various data connectors during research execution.

What is the best way to synthesize findings from academic sources and code repositories?▼

The best way to synthesize findings is by fanning out parallel subagents to gather data from academic sources, code repositories, and the web, then compiling the results into a summary with aggregate source counts.

Do I need an approved research plan before running data collection?▼

Yes, you need an approved research plan before execution, as this Skill runs the actual research, gathers data from configured connectors, and synthesizes the information for counter-review.

When should I not use parallel subagents for information retrieval?▼

You should avoid using parallel subagents for information retrieval when your research plan is not yet approved, as this Skill is designed to execute pre-defined plans and aggregate findings from multiple data sources.

Related Skills

Explore More Education & Research Skills→
644408071-design644408071-design
1

research-goal

Define research goals, track criteria, and compile evidence for financial research.

Community
Intermediate
anserIndicusanserIndicus
1

research-report-generator

Generate HTML and Markdown research reports from community feedback data.

Community
Advanced
hajrasarwar11hajrasarwar11

query-integration-data

Query and modify records across external services and data warehouses.

Community
Advanced
indykishindykish
2

oracle

Bundle prompts and files for second-model API or browser review.

Community
Advanced