Show HN: Runner – the anti-vibe coding agent

18 pointsposted 12 hours ago
by zmccormick7

5 Comments

glial

10 hours ago

Interesting, will check it out.

A persistent issue I have with Cursor et al. is that they hallucinate function arguments when using a function or method from a library. It seems like automatically pulling the library's documentation into the context would be helpful, but I haven't found any tool that does this automatically. Is there any chance that Runner does this?

It doesn't seem like this was the problem you were trying to solve, but reliable use of libraries and APIs is a critical problem to solve if you want LLM-generated code to work.

zmccormick7

9 hours ago

I agree that's a major problem. It's not something I've solved yet. I suspect a web research sub-agent is likely what's needed, so it can pull in up-to-date docs for whatever library you need to work with.

android521

an hour ago

Tried downloading it from my M2 macbook pro 2 times and failed

MuffinFlavored

9 hours ago

> You will need at least a Gemini API key, and for best performance also an OpenAI API key.

Could you explain why both / why not also Claude (why not all three?)

zmccormick7

9 hours ago

Gemini is required for the context management sub-agent. You can use any of OpenAI, Anthropic, or Gemini for the main planning and coding agents, but GPT-5 performs the best in my experience. Claude 4 Sonnet works well too, but it's about twice as expensive.