auto-coder

Implement spec-driven development tasks with tests and atomic commits.

Updated Apr 28, 2026
One-click install
npx skills add https://github.com/yao9906/visrag --skill auto-coder-yao9906
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: auto-coder
Source: https://github.com/yao9906/visrag/tree/main/.claude/skills/auto-coder
Command: npx skills add https://github.com/yao9906/visrag --skill auto-coder-yao9906

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill removes the manual overhead of spec-driven development by turning a project schedule into an executable coding workflow. It helps you identify the next task, implement the required changes, verify them with tests, and record progress consistently.

Core Features & Use Cases

  • Spec synchronization: Keeps chapter-based reference files aligned with the project specification so the skill always works from the latest task map.
  • Task selection and implementation: Finds the next pending task, follows the project architecture, and updates the right files with code that matches the spec.
  • Testing and recovery loops: Runs relevant tests, applies up to three auto-fix rounds, and stops only when the workflow is genuinely blocked.
  • Progress persistence: Updates task status and prepares an atomic commit so development progress is not lost.
  • Use case: A developer wants to finish the next feature in a modular Python repository without manually coordinating spec reading, coding, testing, and status updates.

Quick Start

Ask the auto-coder skill to implement the next pending task in the project spec and complete the code, test, and progress update workflow end to end.

Frequently Asked Questions about auto-coder

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

FAQPage Schema
How do I automate spec-driven development for a modular Python repository?

Spec-driven development automation turns a project schedule into an executable workflow by syncing DEV_SPEC.md, selecting pending tasks, implementing code, running tests, and recording atomic commits.

Can I automatically run pytest and apply auto-fix rounds for new features?

Yes, autonomous coding workflows run relevant pytest tests and apply up to three auto-fix rounds. The loop stops only when the workflow is genuinely blocked by unresolved test failures.

What is the best way to turn pending project tasks into code and atomic commits?

Using an autonomous coding loop converts pending tasks into code by following project architecture, updating files to match specs, and preparing atomic commits to persist development progress.

Does autonomous coding work with chapter-based reference files and DEV_SPEC.md?

Yes, spec synchronization keeps chapter-based reference files aligned with DEV_SPEC.md so the workflow always operates from the latest task map before selecting the next implementation.

How many auto-fix rounds are applied when tests fail during task implementation?

The workflow applies up to three auto-fix rounds when tests fail during task implementation. It stops attempting fixes only when the coding workflow becomes genuinely blocked.

Do I need a DEV_SPEC.md file to start autonomous task scheduling and implementation?

Yes, a DEV_SPEC.md file is required. The workflow syncs this project specification into chapter references to identify the next pending task and implement the required code changes.