auto-coder

Implement the next pending task from a project specification and run tests.

Updated Mar 9, 2026

Search Tags:#pytest #task automation #autonomous coding #spec-driven development #project sync #git commits

Ready to run

npx skills add https://github.com/zhunikaixin297/Enterprise-Modular-RAG-MCP --skill auto-coder-zhunikaixin297

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: auto-coder
Source: https://github.com/zhunikaixin297/Enterprise-Modular-RAG-MCP/tree/main/.claude/skills/auto-coder
Command: npx skills add https://github.com/zhunikaixin297/Enterprise-Modular-RAG-MCP --skill auto-coder-zhunikaixin297

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

It removes the manual overhead of reading a spec, choosing the next task, implementing the code, testing it, and recording progress, so development can move from requirements to a working change with minimal hand-holding.

Core Features & Use Cases

Spec Synchronization: Keeps the project spec split into chapter-based reference files so implementation work always follows the latest documented plan.
Autonomous Task Execution: Finds the next pending task, or a specific task ID when requested, then implements the required code with the repository’s architecture and patterns.
Built-in Validation Loop: Runs relevant tests, retries fixes up to three times, and preserves progress with atomic commit-ready updates.
Use Case: Use it when you want an AI agent to carry a feature from design notes to tested code without micromanaging each step.

Quick Start

Use the auto-coder skill to sync the spec, implement the next pending task, run the necessary tests, and prepare a progress summary for commit.

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 from planning to pytest execution?▼

Spec-driven development automation syncs chapter-based reference files from your project spec, selects the next pending task, implements the code, and runs pytest validation with up to three auto-fix rounds before persisting atomic progress updates.

Can I run autonomous coding on a specific task ID instead of the next pending one?▼

Yes, autonomous coding supports targeted task IDs. When requested, the system bypasses automatic next pending task selection and directly implements the specified task ID, following the same deterministic spec synchronization and iterative development process.

What is the spec synchronization process for autonomous repository workflows?▼

Spec synchronization keeps the project spec split into chapter-based reference files. This deterministic process ensures implementation work always follows the latest documented plan, parsing schedules and coordinating iterative development across code and tests.

How many auto-fix rounds does pytest execution attempt before failing?▼

Pytest execution attempts up to three auto-fix rounds during the built-in validation loop. The system runs relevant tests, retries fixes iteratively, and preserves progress with atomic commit-ready updates if the tests continue to fail after the third round.

Do I need git commits to be prepared manually after autonomous task execution?▼

No, git commits do not require manual preparation. After implementing the required code and running the validation loop, the system automatically prepares a progress summary and atomic commit-ready updates for the repository workflow.