summary: “AI coding agents are autonomous or semi-autonomous AI systems that assist in code writing, debugging, and maintenance through natural language interactions with repository context.” updated: 2026-04-14 group: coding-agents-dev-workflows
AI coding agents
Autonomous or semi-autonomous systems that use AI to assist in writing, debugging, and maintaining code through natural language interactions with repository context.
Key findings
- A 2026 ETH Zurich study, Evaluating AGENTS.md: Are Repository-Level Context Files Helpful for Coding Agents?, found that repository-level context files (like
AGENTS.mdandCLAUDE.md) can degrade AI coding agent performance, contradicting common industry practice of using them for guidance. (Source: Theo, watch?v=GcNu6wrLTJc%7CYouTube) - Anthropic’s workflow for long-running Claude ses
2026 04 14 AI can work worse with Claudemd and agentsmd files Channel Theo
Source Notes
- 2026-04-14: [[lab-notes/2026-04-14-Optimizing-AI-Costs-and-Privacy-with-Local-Open-Source-Models-and-Hybr|“But OpenClaw is expensive…“]]
- 2026-04-07: Claude Code 2.0 MASSIVE Upgrade! (Game Changer)
- 2026-04-08: Claude Code 2.0 MASSIVE Upgrade! (Game Changer)
- 2026-04-08: [[lab-notes/2026-04-08-Claude-Obsidian-Integration-Creating-a-Persistent-AI-Operating-System|Claude + Obsidian = Full AI Operating System]]
- 2026-04-26: [[lab-notes/2026-04-26-Karpathys-AutoResearch-An-AI-Agent-for-Independent-LLM-Program-Improvement|Karpathy’s AutoResearch: An AI Agent for Independent LLM Program Improvement]]