Spent two hours installing a tool to make my coding agent smarter. Then it refused to use it.
Summary
A developer spent two hours installing a tool to improve a coding agent's code reading capabilities, but the agent continued to default to grep despite the superior tool, highlighting the difficulty of changing an agent's established habits.
Similar Articles
@bentlegen: to make *actually* good software, you have to use it a lot but many people are spending more time in their coding agent…
Developer Ben Tlegen observes that building truly good software requires extensive personal use, but warns that many developers are now spending more time inside AI coding agents than in the actual applications they are creating.
@_overment: f-ck me. looks like in 24 hours I’ve created a tool that’s made all coding agents obsolete for me. still iterating, but…
The author claims to have built a tool in 24 hours that makes existing coding agents obsolete, using deterministic hooks and dynamic triggers, and argues that current approaches like Claude Code are flawed.
@_philschmid: Another Bitter Lesson is coming. You might have noticed your coding agents preferring bash over dedicated tool more and…
The post discusses how AI coding agents are increasingly preferring bash scripting over dedicated tools, becoming highly proficient in it, which could lead to another 'Bitter Lesson' in AI development.
I gave my agent a search tool and it declined to use it, plus a few other things I got wrong
The author details building a personal AI agent for book recommendations, sharing lessons on tool compliance, cost management, and feedback loops using a Go-based implementation.
AI agent hit a bug it had predicted 20 minutes earlier-in its own self-written code
A custom AI agent autonomously built a tool, audited its own code, predicted a bug, encountered it later, and recovered by adapting its approach, showcasing advanced self-improvement capabilities.