Enforce explicit assumptions and minimal edits in LLM-generated code.
npx skills add https://github.com/BrunoAMSilva/my-config --skill guidelines-brunoamsilva
Please help me install this Agent Skill. Skill: guidelines Source: https://github.com/BrunoAMSilva/my-config/tree/main/coding/skills/guidelines Command: npx skills add https://github.com/BrunoAMSilva/my-config --skill guidelines-brunoamsilva
These behavioral guidelines reduce common mistakes when LLMs write, review, or refactor code by surfacing assumptions, avoiding overcomplication, and defining verifiable success criteria so changes are safe and reviewable.
When reviewing or implementing code from an LLM, state assumptions, choose the simplest solution, make only surgical edits, and define one or two verifiable success checks.
High-intent search queries and answers about installing and using this skill.
To prevent overcomplication in LLM generated code, enforce simplicity-first implementations that favor minimal code directly solving the request. State explicit assumptions, avoid speculative features, and make surgical changes limited only to what the task requires.
LLM refactoring often introduces unnecessary changes when surgical edits are not enforced. To stop this, limit modifications to only what the task requires, avoid speculative features, and remove only newly orphaned artifacts to keep changes safe and reviewable.
The best way to review LLM code for pull requests is to require explicit assumptions, choose the simplest solution, and define verifiable success checks like tests. Ensure edits are surgical and minimal to produce safe, reviewable results.
To verify LLM bug fixes without breaking functionality, convert tasks into verifiable steps with checks like reproducible tests. Define one or two success criteria to validate completion, ensuring the changes are safe and do not introduce speculative features.
Yes, you can force an LLM to state assumptions before writing code by applying behavioral guidelines. Require it to surface uncertainties and alternative interpretations beforehand, ensuring goal-driven execution and verifiable success criteria for testable results.
You should not use automated LLM code generation without explicit guidelines if the task requires strict surgical edits or verifiable success criteria. Avoid it when simplicity-first implementations are mandatory and unchecked speculative features could compromise the codebase.
ToXMonEnforce conservative coding practices and surgical edits for LLM-generated changes.
MatheusrlrGuide AI coding assistants through code writing, modification, and review processes.
sputnikusApply behavioral guidelines for AI coding assistants to reduce common LLM mistakes.
wangdaqian08Enforce minimal, surgical, and verifiable code changes with explicit tradeoffs.
