skills.rest
Lambenthan
LambenthanCOMMUNITY
Explore all skills by this author →
GitHub

/refine

Iteratively review and edit wiki-linked research artifacts until target quality scores are met.

77|16|Updated May 9, 2026
Search Tags:#refinement#academic writing#research workflow#iterative review#evidence gaps#wiki updates#score optimization
One-click install
npx skills add https://github.com/Lambenthan/empiricalwiki --skill refine-lambenthan
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: /refine
Source: https://github.com/Lambenthan/empiricalwiki/tree/main/.claude/skills/refine
Command: npx skills add https://github.com/Lambenthan/empiricalwiki --skill refine-lambenthan

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It solves the problem of getting research deliverables (ideas, experiment plans, paper drafts) to a high quality standard through repeated review and targeted fixes rather than one-shot editing.

Core Features & Use Cases

  • Multi-round review-to-repair loop: Runs a review, extracts structured issues and actionable items, applies fixes, then re-runs review until the artifact meets a target score.
  • Wiki-aware updates: Updates the relevant EmpiricalWiki pages (ideas, experiments, claims, outputs, and related graph artifacts) based on what the review explicitly flags.
  • Evidence-and-method gap handling: Distinguishes between fixes the model should directly apply versus gaps requiring external actions, and records unresolved items with suggested next steps.
  • Progress reporting: Produces an audit-like REFINE_REPORT with score trajectory, fixed issues, wiki changes, and unresolved issues.

Quick Start

Run refine on your paper draft by invoking: use Skill "refine" with the argument path to your wiki output file, and set a target score with --target-score 8.

Frequently Asked Questions about /refine

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

FAQPage Schema
How do I iteratively improve my academic paper draft until it meets a target quality score?▼

Iterative academic writing refinement runs a multi-round review-to-repair loop on your paper draft, extracting structured issues and applying targeted edits until the artifact reaches your specified target score.

What is the best way to identify and resolve evidence gaps in research experiment plans?▼

Evidence gap handling distinguishes between fixes applied directly to experiment plans and unresolved gaps requiring external actions, recording suggested next steps for method clarity and evidence strength.

How do I run a multi-round review and repair loop on wiki-linked research artifacts?▼

Run a review-to-repair loop by invoking the refinement process with a path to your wiki output file, which applies targeted fixes and re-runs reviews until convergence checks pass a target score.

Does iterative research workflow refinement automatically update EmpiricalWiki entities?▼

Yes, wiki-aware updates automatically modify the relevant EmpiricalWiki pages, including ideas, experiments, claims, and outputs, based on what the review explicitly flags during the refinement process.

Can I track the score optimization trajectory of my research artifact across multiple review rounds?▼

Progress reporting produces an audit-like REFINE_REPORT detailing the score trajectory, fixed issues, wiki changes, and unresolved items upon verdict-based termination of the iterative review.

When should I avoid using automated score-based stopping for academic writing refinement?▼

Avoid automated stopping when your research artifact contains complex evidence gaps requiring external data collection, as the process relies on verdict-based termination and target-score convergence for method clarity.

Related Skills

Explore More Education & Research Skills→
LambenthanLambenthan
77

/review

Review research artifacts and produce structured scores, weaknesses, and improvement plans.

Community
Advanced
duany049duany049

/refine

Iteratively revise research artifacts until a target review score is reached.

Community
Advanced
skyllwtskyllwt
1.6k

exp-eval

Evaluate completed experiment results and update linked idea status and knowledge graph edges.

Community
Advanced
corbat-techcorbat-tech
1

coco-fix-iterate

Automates iterative code refinement cycles combining ML and AI technologies with a focus on quality, safety, and efficiency.

Official
Advanced