agentic-engineering

Implement and validate software changes with eval-first, decomposition-driven workflows.

Updated Sep 13, 2025

Search Tags:#code generation #software quality #regression testing #model routing #agentic #eval-first #cost discipline

Ready to run

npx skills add https://github.com/llmh333/employee_management_spring --skill agentic-engineering-llmh333

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: agentic-engineering
Source: https://github.com/llmh333/employee_management_spring/tree/main/.gemini/skills/agentic-engineering
Command: npx skills add https://github.com/llmh333/employee_management_spring --skill agentic-engineering-llmh333

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Reduces the risk and rework of AI-assisted coding by enforcing an eval-first workflow, careful decomposition, and cost-aware model routing.

Core Features & Use Cases

Eval-first execution loop: Define evaluation gates, capture baseline failure signatures, implement changes, then re-run evals to verify deltas.
Agent-sized decomposition: Break work into independently verifiable units with a clear done condition and a dominant risk per unit.
Model routing and session strategy: Use smaller models for classification/boilerplate, larger models for deep implementation and architecture, while managing when to keep or reset context for reliability.
Quality review guardrails: Focus human review on invariants, edge cases, security/auth assumptions, and rollout coupling rather than style-only differences.

Quick Start

Tell the agent to implement the feature “Add optimistic locking to the Employee update flow” and require an eval-first plan with regression checks and explicit completion criteria before writing any code.

Frequently Asked Questions about agentic-engineering

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

FAQPage Schema

What is an eval-first workflow for AI code generation?▼

An eval-first workflow for AI code generation requires defining evaluation gates and capturing baseline failure signatures before implementation. After changes are made, re-running evals verifies deltas to reduce rework and ensure software quality.

How do I use model routing for agentic software engineering?▼

Model routing for agentic software engineering assigns smaller models to classification or boilerplate tasks, while reserving larger models for deep implementation and architecture. It manages context resets to balance cost discipline with execution reliability.

How do I break down refactoring tasks for agentic engineering?▼

Break down refactoring tasks into agent-sized, independently verifiable units. Each unit must have a clear completion condition and a dominant risk, allowing the agentic engineering process to implement and validate changes safely.

What should human code review focus on when using AI for software development?▼

Human code review during AI software development should focus on architectural invariants, security assumptions, and rollout coupling. It bypasses style-only differences to prioritize edge cases and critical system invariants.

Can I manage AI coding costs and retries for multi-file architectural changes?▼

Yes, agentic engineering workflows manage costs for multi-file architectural changes through measured retries, wall-clock time tracking, and explicit task routing across model tiers to prevent runaway token usage.

When should I not use agentic engineering workflows?▼

Agentic engineering workflows may be unnecessary for trivial edits lacking clear completion criteria. If a task cannot be decomposed into independently verifiable units with regression gates, the overhead of eval-first planning offers little value.