GitHub Repository

Book knowledge distilled into AI agent skills — Clean Code, DDD, Effective Kotlin, and more

6 stars

Using classic dev books to guide AI agents

Name: Using classic dev books to guide AI agents
Availability: InStock
Author: ZLStas

by ZLStas·Feb 21, 2026·3 points·2 comments

Visit Project View on HN

AI Analysis

●MidBig Brain

Book principles as AI-agent prompts, but needs a working workflow to prove value.

Strengths

•Novel angle: codifying engineering wisdom from established texts into agent-consumable format.
•Thoughtful structure with evals/tests and modular skill design suggests genuine system thinking.
•Addresses real problem of maintaining code quality standards across AI-assisted generation.

Weaknesses

•Still experimental—author acknowledges unclear how to wire this into practical workflows (review layer vs. refactoring tool).
•No evidence of production usage or measurable impact on code quality; mostly a framework looking for a use case.

Post Description

I've been experimenting with turning principles from classic software engineering books (Clean Code, DDIA, etc.) into structured "skill" files that AI agents can use during code review. Each skill is an opinionated instruction set grounded in known engineering wisdom — not a summary or excerpt. Repo: https://github.com/ZLStas/skills I'm trying to figure out the best way to wire this into a practical workflow — whether as a review layer or as a tool to iteratively refactor a legacy codebase into something clean and well-structured. A few open questions I'd love input on:

Does it make sense to use book-based principles as a structured lens for AI-driven code review? How would you set up sub-agents to iteratively review LLM output — one agent creates, another evaluates — without the review becoming shallow or repetitive? Has anyone tried a different approach that worked better? How do you maintain project context across multiple review passes so the agent doesn't lose sight of the bigger picture?