Keeping review honest when code is cheap
Eight situations where reading every line is impossible, and what visualdiff does in each.
- Reviewing pull requests written by agentsAgents open pull requests faster than anyone can read them. Read the 10% that decides whether it is safe.
- Large refactors that hide a behaviour changeA refactor is supposed to change nothing. Find the one line that does.
- Stopping hallucinated dependenciesSlopsquatting: attackers register the package names models invent. Catch them at review.
- Seeing which running systems a change reachesThe diff says rds.tf. The map says orders-db, the API in front of it, and the queue behind it.
- Knowing whether review is keeping upThe sanity index: is the codebase getting harder to review as more of it is written by machines?
- An audit trail for regulated teamsEvidence that a person reviewed what mattered, and which code a model wrote.
- Reviewing code you did not write in a system you do not knowA reading order: what is depended on before what depends on it.
- Agents that check their own work firstLet the agent call visualdiff before it opens the pull request, and fix what it finds.