ai-problems-detection

You can prevent AI coding errors by enforcing a mandatory pre-implementation self-diagnostic protocol that evaluates overengineering, duplication, re-inventing existing solutions, missing documentation, and monolithic structure before code execution or commit.

API verification guardrails require confirming method and function existence against the project's installed dependency versions and official documentation before proceeding with any code execution or commit.

Run a defined five-check workflow that evaluates simplicity, reuse, documentation knowledge, duplication risk, and modularity to catch overengineering and missing documentation during feature planning or code modification.

Yes, the protocol applies to reviewing code changes before execution or commit, adding security-focused cues to detect credential exposure, unsafe query patterns, and sensitive-data logging tendencies alongside structural anti-patterns.

Yes, the self-diagnostic workflow includes security-focused review cues that detect credential exposure, unsafe query patterns, and sensitive-data logging tendencies alongside structural anti-patterns before code execution.