hyper-research

Run parallel Codex and Claude research paths to produce structured pre-implementation reports.

3|Updated May 10, 2026
One-click install
npx skills add https://github.com/zeikar/hyperclaude --skill hyper-research
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: hyper-research
Source: https://github.com/zeikar/hyperclaude/tree/main/skills/hyper-research
Command: npx skills add https://github.com/zeikar/hyperclaude --skill hyper-research

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes assets (resource) components.

What problem does it solve?

hyper-research prevents costly design rework by gathering prior art, pitfalls, and recommendations before you start implementing a non-trivial change.

Core Features & Use Cases

  • Parallel research artifacts: Runs Codex research and Claude-native research in parallel by default to produce two reviewable trace files under .hyperclaude/research/.
  • Explicit single-path option: Supports Codex-only or Claude-only research when the user explicitly requests it, reducing wasted time when one perspective is sufficient.
  • Plan-ready output: Produces findings intended to be integrated into the next planning step, with consistent slug conventions for downstream workflow matching.

Quick Start

Use hyper-research when you are about to start a substantial new task and want prior-art and pitfalls before designing the plan.

Frequently Asked Questions about hyper-research

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

FAQPage Schema
How do I gather prior art and pitfalls before implementing a new software feature?

Pre-implementation research gathers prior art, pitfalls, and recommendations before you start coding. It runs parallel research paths to produce structured artifacts, reducing costly design rework and implementation surprises for non-trivial tasks.

When should I run pre-implementation research for agentic workflows?

Pre-implementation research is necessary for non-trivial features, architectural changes, and any work where you need prior art and pitfalls analysis before planning. It prevents design mistakes by gathering recommendations before substantial new tasks.

Can I run Claude research and Codex research in parallel for the same task?

Yes, parallel research artifacts are generated by default, running Claude-native and Codex research paths simultaneously. You can also explicitly choose a single-path option with either tool when one perspective is sufficient.

What is the best way to structure pre-implementation research findings for planning?

The best way to structure research findings is using a Prior Art, Pitfalls, Recommendations, and Open Questions report format. This plan-ready output is written into trace files with consistent slug conventions for downstream workflow matching.

Does pre-implementation research work with Claude Code for architectural changes?

Yes, pre-implementation research works with Claude Code to reduce design mistakes for architectural changes. It produces reviewable trace files under a dedicated research directory, enforcing a structured report format for integration into the next planning step.