skills.rest
SpillwaveSolutions
SpillwaveSolutionsOFFICIAL
Explore all skills by this author →
WebsiteGitHub

architect-agent

Coordinate AI agents through structured delegation, grading, and iterative improvement.

19|8|Updated Oct 27, 2025
Search Tags:#project management#quality assurance#code review#software development#AI agent#delegation#human-in-the-loop
One-click install
npx skills add https://github.com/SpillwaveSolutions/architect-agent --skill architect-agent
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: architect-agent
Source: https://github.com/SpillwaveSolutions/architect-agent/tree/main/.
Command: npx skills add https://github.com/SpillwaveSolutions/architect-agent --skill architect-agent

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

AI coding assistants often work autonomously, lacking human oversight, structured delegation, and quality control. This skill provides a robust framework for developers to maintain control, ensure quality, and track progress when delegating tasks to AI agents. It transforms a simple AI prompt into a managed, auditable development process.

Core Features & Use Cases

  • Structured Delegation: Create detailed, multi-phase instructions for AI coding agents, ensuring clarity, consistency, and alignment with project goals.
  • Quality Assurance & Grading: Rigorously review and grade AI-generated code against clear rubrics, providing a feedback loop for continuous improvement and iterative refinement until quality thresholds are met.
  • Comprehensive Audit Trails: Maintain real-time logs of all AI decisions, commands, and changes, offering full visibility, traceability, and compliance documentation.
  • Use Case: A lead developer needs to implement a complex feature. Instead of just prompting an AI, they use this skill to define precise instructions, track the AI's progress through detailed logs, and rigorously grade the final code, ensuring it meets high standards before integration.

Quick Start

To begin, tell the AI: This is a new architect agent, help me set it up.

Frequently Asked Questions about architect-agent

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

FAQPage Schema
How do I set up an AI agent to manage code review and quality assurance with human oversight?▼

This skill transforms Claude Code into a specialized architect agent that plans, delegates tasks to code agents, and grades output against a structured rubric. It establishes human-in-the-loop control, audit trails, and iterative quality improvement until code meets your ≥95% quality threshold.

What's the best way to delegate software development tasks to an AI agent while maintaining quality control?▼

Define precise, multi-phase instructions using this skill's structured delegation framework, then rigorously grade AI-generated code against a 100-point rubric. The skill tracks all decisions and changes in real-time logs, enabling feedback loops for continuous refinement until quality targets are met.

How do I create detailed instructions for AI coding agents to follow during project implementation?▼

This skill provides a templated automation framework for instruction generation that ensures clarity and alignment with project goals. It covers workspace initialization, delegation parameters, and grading criteria, enabling consistent execution across multiple AI agents and cross-workspace collaborations.

Can I use an AI agent for code review and maintain an audit trail of all changes?▼

Yes. This skill maintains comprehensive audit trails logging all AI decisions, commands, and code changes in real time. Combined with its grading rubric and iterative feedback loop, you gain full visibility and compliance documentation while the agent reviews and improves code quality.

What's the difference between autonomous AI coding and delegating with structured quality control?▼

Autonomous AI agents lack oversight and verifiable quality gates. This skill adds structured delegation, rigorous grading against explicit rubrics, human decision points, and memory management across workflows—transforming an untracked process into an auditable, managed development cycle.

Do I need prior experience with AI agents to use structured delegation and grading?▼

No. The skill uses a progressive disclosure architecture and templated framework designed for lead developers new to AI-driven workflows. Start by telling the agent 'This is a new architect agent, help me set it up' to initialize workspace, instructions, and grading parameters.

Related Skills

Explore More Software Engineering Skills→
matthew-johnsonmatthew-johnson

opencode

Delegate coding tasks to an autonomous OpenCode CLI worker.

Community
Advanced
thompson0012thompson0012
28

design-compounder

Extracts artifact-backed design lessons from completed sprint outcomes for cross-sprint knowledge updates.

Community
Advanced
mhintz1980mhintz1980

evaluation

Evaluate agent systems with multidimensional rubrics and automated or human modes.

Community
Advanced
nwiizonwiizo
147

HITL Approval Workflow

Manage human-in-the-loop approval workflows for agent operations in ccswarm.

Community
Advanced