skills.rest
thangvawn
thangvawnCOMMUNITY
Explore all skills by this author →
GitHub

agentic-engineering

Automate engineering workflows with AI agents and human oversight.

Updated Apr 6, 2026
Search Tags:#quality control#workflow automation#AI agents#code execution#agentic engineering
One-click install
npx skills add https://github.com/thangvawn/agent_financial --skill agentic-engineering-thangvawn
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agentic-engineering
Source: https://github.com/thangvawn/agent_financial/tree/main/.cursor/skills/agentic-engineering
Command: npx skills add https://github.com/thangvawn/agent_financial --skill agentic-engineering-thangvawn

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill addresses the complexity of engineering workflows by leveraging AI agents for execution, with humans in the loop for quality and risk control.

Core Features & Use Cases

  • Eval-First Execution: Ensures quality by defining completion criteria, decomposing tasks, and measuring outcomes before implementation.
  • Decomposition: Breaks down work into manageable units with clear done conditions and single dominant risks.
  • Model Routing: Routes tasks based on complexity to appropriate model tiers (Haiku, Sonnet, Opus).
  • Session Strategy: Manages sessions for closely-coupled units, resets for phase transitions, and compresses after milestones.
  • Review Focus: Prioritizes invariants, error boundaries, security, and hidden coupling in AI-generated code.
  • Cost Discipline: Tracks per-task metrics like model, token estimate, retries, and wall-clock time to optimize resource usage.

Quick Start

Activate the agentic-engineering skill to begin a new project with eval-first execution and AI agent assistance.

Frequently Asked Questions about agentic-engineering

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

FAQPage Schema
How do I automate complex engineering workflows with AI agents while maintaining quality control?▼

Automating complex engineering workflows with AI agents involves using eval-first execution, model routing, and session management, keeping humans in the loop for risk and quality control throughout the code execution process.

What is eval-first execution in AI agent task automation?▼

Eval-first execution in AI agent task automation ensures quality by defining completion criteria, decomposing tasks into manageable units with clear done conditions, and measuring outcomes before actual implementation begins.

How does model routing work for AI agents in software development?▼

Model routing for AI agents in software development assigns tasks based on complexity to appropriate model tiers like Haiku, Sonnet, or Opus, optimizing resource usage and ensuring efficient task execution.

Do I need to define structured tasks before using AI agents for workflow automation?▼

Yes, using AI agents for workflow automation requires a structured task definition to enable proper task decomposition, model routing, and effective session management for your software development project.

What are the limitations of using AI agents for code execution in engineering projects?▼

Using AI agents for code execution requires strict human oversight to review invariants, error boundaries, security, and hidden coupling in AI-generated code, while tracking per-task metrics like token estimates and retries to maintain cost discipline.

Related Skills

Explore More Software Engineering Skills→
oabdelmaksoudoabdelmaksoud
3

agentic-engineering

Decompose AI engineering tasks into verifiable units with eval-first execution.

Community
Advanced
heymishyheymishy

feature-review

Validate feature review decisions against governance criteria for traceability and decision integrity.

Community
Basic
sourcegraphsourcegraph
32

Evaluate Benchmark Traces

Evaluate AI agent benchmark traces for data integrity, output quality, and efficiency.

Official
Advanced
jfrometajfrometa

plan-writing

Decompose complex software work into small, verifiable tasks with dependencies.

Community
Basic