research-pipeline

Automate research from idea discovery to manuscript generation with configurable gates.

Updated Apr 29, 2026
One-click install
npx skills add https://github.com/jkfee/Auto-Research --skill research-pipeline-jkfee
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: research-pipeline
Source: https://github.com/jkfee/Auto-Research/tree/main/skills/research-pipeline
Command: npx skills add https://github.com/jkfee/Auto-Research --skill research-pipeline-jkfee

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates the entire end-to-end research lifecycle from idea discovery to manuscript generation, reducing manual overhead and accelerating scientific progress.

Core Features & Use Cases

  • Chains idea discovery, implementation, deployment, auto-review, and optional paper writing into a single workflow.
  • Supports configurable gates (AUTO_PROCEED, HUMAN_CHECKPOINT, AUTO_WRITE) and venues, enabling end-to-end automation for multiple research topics.
  • Outputs structured artifacts such as IDEA_REPORT.md, NARRATIVE_REPORT.md, and AUTO_REVIEW.md to enable traceability and reproducibility.

Quick Start

Initiate the full end-to-end research pipeline for a given topic by supplying a direction.

Frequently Asked Questions about research-pipeline

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

FAQPage Schema
How do I automate the research lifecycle from idea discovery to paper writing?

Automate the end-to-end research lifecycle by configuring gates like AUTO_PROCEED, HUMAN_CHECKPOINT, and AUTO_WRITE to orchestrate idea discovery, experiment implementation, auto-review, and manuscript generation.

What is the best way to structure an automated research pipeline for multiple topics?

Structure an automated research pipeline by chaining idea discovery, implementation, and auto-review loops into a single workflow, outputting structured artifacts like IDEA_REPORT.md and NARRATIVE_REPORT.md for traceability.

Can I set human checkpoints during an automated experiment and paper-writing process?

Yes, you can set human checkpoints during automated experiment and paper-writing processes by configuring the HUMAN_CHECKPOINT gate to pause the workflow for manual validation before proceeding.

How does an auto-review loop work when running experiments autonomously?

An auto-review loop evaluates experimental results autonomously by applying configurable REVIEWER_DIFFICULTY settings, iterating until conditions are met before advancing to the paper-writing stage.

Do I need to configure venue settings before generating a manuscript automatically?

Yes, you need to configure venue settings before generating a manuscript automatically, as the VENUE parameter tailors the paper-writing stage to specific publication requirements.

What limitations exist when using autonomous planning for end-to-end research automation?

Limitations of autonomous research automation include the need for explicit gate configurations like ARXIV_DOWNLOAD and AUTO_PROCEED, requiring manual setup to ensure proper experiment execution and manuscript reproducibility.