skills.rest
Manvendra08
Manvendra08COMMUNITY
Explore all skills by this author →
GitHub

agentic-engineering

Enforce eval-first execution and cost-aware model routing for AI agent engineering workflows.

1|Updated May 12, 2026
Search Tags:#task decomposition#agentic engineering#cost optimization#model routing#ai-assisted development#eval-first
One-click install
npx skills add https://github.com/Manvendra08/TradingBot --skill agentic-engineering-manvendra08
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agentic-engineering
Source: https://github.com/Manvendra08/TradingBot/tree/main/_agent/skills/agentic-engineering
Command: npx skills add https://github.com/Manvendra08/TradingBot --skill agentic-engineering-manvendra08

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Unstructured AI agent engineering workflows often lead to unreliable outputs, wasted compute resources, and missed quality controls, making it hard to deliver consistent, production-ready software with AI assistance.

Core Features & Use Cases

  • Eval-First Execution: Define capability and regression evals upfront to measure implementation impact and catch regressions before they reach production.
  • Structured Task Decomposition: Break work into 15-minute independent units with clear done conditions to reduce risk and improve verifiability of AI-generated work.
  • Cost-Aware Model Routing: Match model tiers (Haiku, Sonnet, Opus) to task complexity to optimize compute cost without sacrificing output quality.
  • Use Case: When building a new feature for a trading bot, use this skill to decompose the work into testable units, route implementation tasks to Sonnet and architecture reviews to Opus, and track cost per task to avoid overspending.

Quick Start

Use the agentic-engineering skill to plan and execute the next feature implementation for your project with built-in eval checks and cost tracking.

Frequently Asked Questions about agentic-engineering

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

FAQPage Schema
How do I prevent regressions when using AI agents for software development?▼

To prevent regressions in AI agent engineering, adopt an eval-first approach by defining capability and regression evaluations upfront. This measures implementation impact and catches errors before they reach production, ensuring reliable outputs.

What is structured task decomposition for AI-assisted development?▼

Structured task decomposition breaks AI development work into 15-minute independent units with clear done conditions. This reduces risk and improves the verifiability of AI-generated code changes for multi-file software projects.

How do I optimize compute costs when routing AI agent tasks across different model tiers?▼

Cost-aware model routing optimizes compute costs by matching model tiers like Haiku, Sonnet, and Opus to task complexity. Route implementation to Sonnet and architecture reviews to Opus to balance spend and quality.

Can I use eval-first execution for multi-file code refactoring and root-cause analysis?▼

Yes, eval-first execution applies directly to AI agent workflows for implementation, refactoring, root-cause analysis, and multi-file code changes. It enforces measurable delivery, risk mitigation, and quality control.

Why does my AI agent engineering workflow produce unreliable outputs and waste compute?▼

Unstructured AI agent workflows cause unreliable outputs and wasted compute by missing quality controls. Enforcing eval-first execution, structured task decomposition, and cost-aware model routing mitigates these risks.

Related Skills

Explore More Software Engineering Skills→
xu-xiangxu-xiang
1.8k

eval-harness

Define, execute, and report capability and regression evals for Claude Code sessions.

Community
Intermediate
linnefromicelinnefromice

eval-harness

Evaluate Claude Code sessions with code-based, model-based, and human graders.

Community
Advanced
safzanpiranisafzanpirani
1

deslop

Remove redundant and suboptimal code sections from AI-generated source files.

Community
Basic
Zenobia000Zenobia000
2

agentic-engineering

Structure AI agent engineering workflows with evals, task decomposition, and model routing.

Community
Advanced