hf-review

To conduct an independent code review, use subagents or new sessions to evaluate spec, plan, design, and implementation code. This independent assessment maintains objectivity by manually executing tests and checking evidence without developer bias, providing direct findings and conclusions for quality assurance.

Reviewing code integrity across artifacts involves a multi-step process: reading spec, plan, and design files, manually executing associated tests, checking evidence, and providing findings. This ensures all aspects of the implementation are covered for unbiased quality assurance.

To get started with a thorough code review, run the review process directly on your 'spec.md' file. The system requires access to the codebase and associated files to read artifacts, manually execute tests, and output findings and conclusions.

Yes, unbiased code review requires access to the codebase and associated files for full functionality. Accessing these files is necessary to read artifacts, manually execute tests, check evidence, and provide accurate findings and conclusions for the implementation.

The best way to avoid manual review pitfalls is conducting an independent assessment using subagents or new sessions. This approach reads artifacts, manually executes tests, checks evidence, and outputs findings directly to the system, ensuring objectivity and code integrity.