3 ms·
Ask HN: Is it better to have no Agent.md than a bad one?
Please share your real word experiences.
What is a bad one and why?
- verdverm 7mo agoMy approach is to keep them minimal. Make them a table of contents and spark notes, spread them across the repo, keep them up to date as the repo changes. Something is better than nothing. It's pretty hard to have bad ones, also takes effort to craft good ones. The bad ones are the ones with incorrect information.
- parvardegr 7mo agoHmm sounds reasonable
- verdverm 7mo agoIf paired this with system prompts derived from Claude Code. This makes gemini-flash nearly on par with the big boy models. Saves a ton of time and money
- nullcathedral 7mo agoOne approach (Claude Code) is to evolve it over time. Start small and run /insights often and use that to refine the CLAUDE.md as needed. https://github.com/trailofbits/claude-code-config?tab=readme-ov-file#continuous-improvement https://github.com/trailofbits/claude-code-config?tab=readme...
- parvardegr 7mo agoThanks for this
- e5bed97b3349 7mo agoYes, zero instructions are better than wrong instructions because outdated or incorrect context actively forces the AI to write broken code. Start with nothing, and only add a single sentence to your Agent.md after the AI repeatedly makes the same specific mistake.
- DANmode 7mo agoWithout further instruction, it falls back to the default logic.
- allinonetools_ 7mo ago[flagged]
- MidasTools 7mo ago[flagged]
- andrei_says_ 7mo agoHow do you instruct the LLM to use each of these files? Do you use “soul” for example? Or?