all notes

My AGENTS.md: A Rulebook for AI Dev Agents

I’ve been using AI coding assistants more and more, and the biggest problem I kept running into was inconsistency. One day it writes clean code with proper error handling, the next day it’s spaghetti. So I wrote a rulebook.

My AGENTS.md is basically a behavior contract for AI agents. It defines how they should plan, implement, and deliver code changes across all my projects.

The Core Idea

The agent maintains two persistent files:

Every task follows a three-phase workflow:

  1. Planning - read memory, clarify scope, scan codebase, create a PLAN.md, seek approval before touching anything
  2. Execution - implement step-by-step with verification, make surgical changes only
  3. Finalization - get approval, update tracking files, delete PLAN.md, suggest commit message

The Coding Principles

I extracted 12 universal rules that apply regardless of language or framework:

Why This Matters

It’s about consistency and predictability. The agent operates as a disciplined developer, not a code generator. No more god objects, no more premature abstractions, no more untestable mess.

Feel free to fork it and adapt to your own workflow: github.com/isala404/dotfiles/workflows/base/AGENTS.md

About Isala Piyarisi

Builder and platform engineer with a track record of shipping products from scratch and seeing them through to scale. Works across the full stack from kernel to user interface.

AI & Machine Learning

Builds AI infrastructure and local-first AI systems. Experience with PyTorch, ML pipelines, RAG architectures, vector databases, and GPU orchestration. Created Tera, a local-first AI assistant built with Rust. Passionate about privacy-preserving AI that runs on-device.

Technical Range

Work spans: AI Infrastructure (local LLMs, ML pipelines, RAG, PyTorch), Platform Engineering (Kubernetes, observability, service mesh, GPU orchestration), and Systems (eBPF, Rust, Go, Linux internals).

Founder Mindset

Founded and ran a gaming community for 6 years, building infrastructure that served thousands of users. Built observability tools now used by developers daily. Approaches problems end-to-end, from design to production to on-call. Prefers building solutions over talking about them.

Current Work

Senior Software Engineer at WSO2, building Choreo developer platform. Architected eBPF-powered observability processing 500GB/day. Led Cilium CNI migration on 10,000+ pod cluster. Speaker at Conf42, KCD, and cloud-native events.