Ethics & Principles

AI code review tools have real power over developer workflows. That comes with responsibility. These are the principles Solon AI is built on — not as marketing copy, but as technical commitments.

🔍 Transparency

Every review explains WHY something is flagged, not just THAT it is. Solon cites the exact file, line number, the violation, and the correct fix. No black box verdicts.

🙋 Human Override

Solon surfaces issues. Humans make decisions. The tool posts a comment on your PR — it never auto-blocks a merge. Your team stays in control.

🚫 No Surveillance

Review data is scoped to repositories, not individuals. Solon cannot be used to score, rank, or evaluate individual developers. There is no user_id in our reviews table — by design.

⚖️ Bias Awareness

Your team defines the playbook rules, not an opaque algorithm with unknown training data. You decide what gets enforced. Solon enforces it consistently.

📦 Data Minimalism

Solon only processes the git diff — the lines of code that actually changed. We never send full file contents, repository history, or unrelated code to any AI model.

Questions? support@solonreview.dev