quality-gates

Automate quality gates and severity classification for multi-agent workflows.

2|Updated Dec 25, 2025
One-click install
npx skills add https://github.com/tianzecn/myclaudecode --skill quality-gates-tianzecn
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: quality-gates
Source: https://github.com/tianzecn/myclaudecode/tree/main/plugins/orchestration/skills/quality-gates
Command: npx skills add https://github.com/tianzecn/myclaudecode --skill quality-gates-tianzecn

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill introduces quality gates—approval points, iteration loops, and severity classification—to ensure multi-agent workflows produce reliable, user-aligned results before moving forward.

Core Features & Use Cases

  • User approvals: cost gates, quality gates, and final acceptance prompts.
  • Iteration loops: automated refinement until a quality threshold is reached.
  • Severity classification: CRITICAL, HIGH, MEDIUM, LOW to prioritize fixes.
  • Consensus patterns: drive multi-reviewer decisions and feedback loops.

Quick Start

Configure a plan with explicit approval gates and a max-iteration cap; run the workflow and prompt users for final validation before advancing.

Frequently Asked Questions about quality-gates

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

FAQPage Schema
How do I set up approval gates in multi-agent workflows?

Approval gates are checkpoints that require user validation before a workflow advances. Configure explicit approval gates in your plan, set a maximum iteration cap, run the workflow, and prompt users for final validation at each gate to ensure quality thresholds are met before proceeding to the next stage.

What's the best way to implement iterative refinement with feedback loops?

Iterative refinement automates repeated cycles until quality reaches a threshold. Define exit criteria and iteration limits in your workflow configuration, then the system automatically refines output based on feedback until stopping conditions are met or the cap is reached.

Can I use severity classification to prioritize issues across my workflow?

Yes. Severity classification categorizes findings as CRITICAL, HIGH, MEDIUM, or LOW to prioritize fixes and guide multi-agent decisions. This enables consensus patterns where multiple reviewers focus effort on high-impact issues before lower-priority ones.

How do I enforce cost and quality gates in deployment workflows?

Cost and quality gates are approval checkpoints that validate resource spend and quality metrics before deployment. Configure these as approval-driven gates with audit logging and user prompts to ensure compliance and user alignment before advancing through deployment stages.

Do I need test-driven development patterns to use approval gates?

No, but approval gates support TDD-driven loops as an option. You can configure them for any approval-driven, iterative, or feedback-loop workflow across design, development, testing, and deployment—TDD is one supported pattern among many.

What are the limitations of iteration caps in quality gate workflows?

Iteration caps prevent infinite refinement loops by setting a hard limit on cycles. Once the cap is reached, the workflow exits regardless of whether quality thresholds are met, so set caps conservatively to balance thoroughness against cost and time constraints.