LLM Bot that reviews Pull Requests for Apache NuttX RTOS
This PR demonstrates a good understanding of NuttX requirements. Adding specific issue references and detailing the testing environment would further strengthen it.
Remember to define our requirements precisely, mark the boundaries clearly. Otherwise our LLM will wander off, and hallucinate strange new ways to validate our Pull Request.
It’s hard to control the LLM Output, hence we politely asked LLM to tone down the response. (And be a little less irritating)
That’s the law of “LLM See, LLM Do”! We fed the input in Markdown Format, so it produced the output in Markdown Format too.
Gemini API will sometimes fail with HTTP 500 (not sure why). And the Failed Request will be counted in our Daily Quota! (50 per day)
We’re still monitoring? Our Bot seems helpful for newer NuttX Contributors, we noticed better quality in some Pull Requests.
Many Thanks to my GitHub Sponsors (and the awesome NuttX Community) for supporting my work! This article wouldn’t have been possible without your support.
