Design and audit an AI agent harness

Get skill on GitHubRaw SKILL.md on GitHub

Opens the skill on GitHub

Source DenisSergeevitch

Installnpx skills add DenisSergeevitch/agents-best-practices

Submit a skill

Overview

This skill is a provider-neutral guide for designing, auditing, and troubleshooting the system around an AI agent, including its loop, tools, memory, permissions, and prompts. It applies to any domain, from coding to support, finance, and operations, and covers topics such as context compaction, MCP connectors, evals, observability, and safety. You ask it for an MVP blueprint, a review of an existing harness, or an explanation of a pattern. It targets OpenAI, Anthropic, and compatible APIs.

Use cases

Alternatives

  • Add Firebase Authentication to an appThis skill guides adding Firebase Authentication to an app for user sign-in and account management. It covers users, identity providers, and tokens, then steps through provisioning, client setup, and security rules that limit data access by signed-in user.Coding agentFirebaseAuth
  • Agents SDK (Cloudflare)Builds agents on Cloudflare's Agents SDK with Workers-friendly patterns and constraints.Coding agent
  • Algorithmic ArtCreates seeded algorithmic art with p5.js and interactive parameter exploration.Coding agent
  • Animate web pages with GSAPThis skill is GreenSock's guide to the core GSAP API: tweens with to, from, and fromTo, easing, duration, stagger, and defaults. It also covers matchMedia for responsive and reduced-motion animation and when to pick GSAP over CSS or other libraries.Coding agentAnimationFrontend
  • Apple DesignApplies Apple-like interface and physical-motion design patterns to web experiences.Coding agentDesignFrontend