skills.rest
fitlab-ai
fitlab-aiOFFICIAL
Explore all skills by this author →
GitHub

refine-task

Translate code review feedback into structured refinement reports and update task metadata.

84|5|Updated Mar 6, 2026
Search Tags:#automation#refinement#quality-assurance#task-management#workflow-automation#code-review#report-generation
One-click install
npx skills add https://github.com/fitlab-ai/agent-infra --skill refine-task
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: refine-task
Source: https://github.com/fitlab-ai/agent-infra/tree/main/.agents/skills/refine-task
Command: npx skills add https://github.com/fitlab-ai/agent-infra --skill refine-task

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

自动将代码审查反馈转化为结构化的修复报告，并产出 refinement.md 或 refinement-r{N}.md，同时更新任务状态以追踪改动。

Core Features & Use Cases

  • 读取最新审查产物和实现产物，生成 refinement.md 或 refinement-r{N}.md，确保修复可追溯。
  • 按 Blocker > Major > Minor 的优先级处理修复，更新 task.md，并输出清晰的修复报告。
  • 支持工作流扩展：将修复信息同步到 Issue、记录改动日志、提供修复报告模板。

Quick Start

运行 refine-task 处理当前任务的审查反馈，生成 refinement.md，并更新 task.md。

Frequently Asked Questions about refine-task

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

FAQPage Schema
How do I convert code review feedback into a structured refinement report?▼

Handling code review feedback automatically involves reading the latest review and implementation artifacts, validating consistency, and generating a refinement.md file. This workflow processes feedback by Blocker, Major, and Minor priorities while updating task metadata.

How do I track code review fixes and update task status automatically?▼

Tracking code review fixes automatically requires processing review artifacts and updating task.md with the latest refinement status. This ensures traceable refinement artifacts are generated and task progress reflects the resolved review feedback accurately.

What is the process for generating traceable refinement artifacts from review feedback?▼

Generating traceable refinement artifacts from review feedback involves reading the latest implementation and review files, validating their consistency, and producing refinement.md or refinement-r{N}.md files. This creates a clear record of actionable changes for software development workflows.

Does this refinement workflow support prioritizing fixes by severity like Blocker or Major?▼

Yes, this workflow supports prioritizing fixes by severity, processing Blocker, Major, and Minor issues in order. It reads review artifacts to apply these actionable changes and generates a refinement report documenting the fixes.

Can I generate multiple rounds of refinement reports for iterative code reviews?▼

Yes, you can generate multiple rounds of refinement reports for iterative code reviews by producing refinement-r{N}.md files. This allows you to track successive iterations of review feedback processing while maintaining consistent task metadata updates.

Related Skills

Explore More Software Engineering Skills→
TranVanTienDatTranVanTienDat

openspec-verify-change

Verify change implementations against openspec artifacts for completeness and correctness.

Community
Advanced
pxlcrtivpxlcrtiv

to-issues

Decompose strategic plans into individual issues for project tracking.

Community
Intermediate
NeoLabs-SystemsNeoLabs-Systems
23

subagent-driven-development

Dispatch fresh subagents for task execution and two-stage code review.

Official
Advanced
ArvorCoArvorCo
96

checklist

Generate Markdown validation checklists from spec, plan, and tasks documents.

Community
Advanced