byte-reviewlisted

# Byte Review Review the actual artifact and evidence. Choose review dimensions based on what could realistically make the result wrong, harmful, confusing, or incomplete. ## Method - Establish the intended behavior and scope. - Check relevant active lessons for repeated mistakes. - Inspect relevant code, files, runtime behavior, tests, or sources. - Prioritize concrete findings by impact and confidence. - Distinguish confirmed defects from risks, preferences, and missing evidence. - Recommend changes that materially improve the outcome. Possible lenses include product fit, UX, correctness, architecture, maintainability, security, privacy, performance, testing, market evidence, and delivery readiness. Do not simulate a meeting or force every role to comment. For code review, lead with actionable findings tied to precise locations. For a product review, lead with the decision or highest-impact gap. If there are no material findings, say so and note meaningful verification limits. Write a review artifact only when it will be tracked or used for later iteration. Do not require a verdict schema, a fixed number of findings, or a Byte OS state transition. When review evidence confirms a reusable mistake, false completion claim, or requirement misunderstanding, update the matching `.byte-os/LESSONS.md` entry with evidence and a prevention rule, not blame. ## Source And Updates Canonical repository: [elan6666/your-bytedance-skills](https://github.com/elan6666/your-bytedance-