research-refine-pipeline

Chain method refinement with claim-driven experiment planning into a structured research package.

Updated Apr 21, 2026
One-click install
npx skills add https://github.com/Shallow-W/llm-wiki --skill research-refine-pipeline-shallow-w
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: research-refine-pipeline
Source: https://github.com/Shallow-W/llm-wiki/tree/main/.claude/skills/research-refine-pipeline
Command: npx skills add https://github.com/Shallow-W/llm-wiki --skill research-refine-pipeline-shallow-w

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It eliminates the back-and-forth between method refinement and experiment planning by producing a single, coherent end-to-end package: a focused final proposal plus a claim-grounded experiment roadmap.

Core Features & Use Cases

  • End-to-end chaining: composes research-refine and experiment-plan so you get both the stabilized method thesis and the experiments that validate its claims.
  • Stability-first guardrail: prevents planning a large experiment suite on top of an unstable or inconsistent method by using a triage + revision gate.
  • Aligned outputs for iteration: generates a final proposal, review summary, refinement report, experiment plan, experiment tracker, and an integration pipeline summary so you can proceed directly to execution.
  • Use case: when you want the whole workflow “串起来” (tied together) from vague research direction to a finalized proposal and a detailed experimental plan.

Quick Start

Use this skill to run a one-shot pipeline that turns your research direction into a refined method proposal and a detailed experiment plan.

Frequently Asked Questions about research-refine-pipeline

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

FAQPage Schema
How do I generate a research proposal and experiment plan in one run?

To generate a research proposal and experiment plan in one run, use an end-to-end pipeline that chains method refinement with claim-driven experiment planning. This produces a finalized proposal and a detailed experiment roadmap together.

What is the best way to connect method refinement with experiment planning?

The best way to connect method refinement with experiment planning is through a stability-first workflow that gates revisions before expanding experiments. This ensures the experiment suite is built on a stable, validated method thesis rather than an inconsistent proposal.

Can I get a complete experiment roadmap from a vague research direction?

Yes, you can get a complete experiment roadmap from a vague research direction by applying a one-shot pipeline. It performs triage and revision to stabilize your methodology, then automatically expands the validated claims into a structured experiment plan and tracker.

Does the pipeline prevent planning experiments on an unstable methodology?

Yes, the pipeline prevents planning experiments on an unstable methodology by applying a triage and revision gate. This stability-first guardrail blocks the generation of a large experiment suite until the method thesis becomes consistent and focused.

What output files do I get from an end-to-end research refinement workflow?

An end-to-end research refinement workflow outputs a structured set of files: a final proposal, review history, refinement report, experiment plan, experiment tracker, and an integration pipeline summary for direct execution.