Bob Martin publishes coding rules for AI agents
Bob Martin publishes coding rules for AI agents
Uncle Bob’s ub-mode skill sets guardrails for AI coding agents, requiring minimal changes, real tests, and root-cause fixes.
Bob Martin, author of "Clean Code", published a set of rules that guide his own AI agents when writing code.
Key points:
- A check that only hides a failure without addressing its root cause cannot be considered a fix.
- If a test still passes when all functions return nothing, that test should be discarded.
- Code changes should be minimal, but sufficient to solve the task at hand.
- Before modifying code, write a test that exposes the existing bug.
Installing this skill can help prevent AI agents from hiding errors and avoid situations where passing tests create a false impression that the code is working.
Installation prompt:
Install the ub-mode skill from the GitHub repository unclebob/ub-mode into my skills folder and use the /ub-mode command for the next code change. If crapper, dryer, or mutator are not installed, skip the corresponding steps.
Repository: unclebob/ub-mode
Why it matters
AnalysisThe rules keep AI coding agents accountable and reduce false confidence from tests that pass for the wrong reasons.
Discuss in community
Share your questions and insights with developers
Related in Ecosystem
Practical guides, core terminology, and agent skills
Bob Martin Publishes Rules for AI Agents Writing Code
Bob Martin, author of Clean Code, published the ub-mode GitHub repository with rules that guide his own AI coding agents.
Open-source project with 74 skills for game development
A GitHub open-source project called awesome-gamedev-agent-skills brings together 74 skills for Unity, Godot, Unreal Engine and other game engines, with automatic skill selection and support for Claude Code and Codex.
Claude Code Projects: How a Single AI Orchestrates a Team of Autonomous Agents
A deep dive into Claude Code Projects: Coordinator and Threads architecture, parallel branches, shared memory via MEMORY.md, and cloud agent management.
AI Agents (Autonomous Agents)
Software systems based on LLMs that can autonomously perceive the state of the environment, decompose complex goals, invoke external tools, and iteratively correct their own mistakes.
Jev and System One AI: A Complete Practical Guide to the Decision Model from TypeSafe
A deep engineering breakdown of the flagship Jev model: System One architecture versus LLMs, Choice/Score/Noul primitives, request parallelism, confidence mathematics, video use cases, code, and an overview of local open-weight alternatives (Laya, GLiNER2.5, CLM-8B).
AI Code Smells
Typical antipatterns and markers of synthetic code ('code smells' from AI). Excessive obvious comments, utility duplication in every file, fake mocks, and nonsensical try-catch blocks that indicate unchecked generation.