Tests
The Tests section documents the protocols Better Robots.txt uses to observe outcomes rather than assert them. A test route exists for each protocol so that the methodology can be cited directly when a result is published or contested.
Why dedicated test routes exist
Visibility outcomes (search appearance, AI answer mentions, archive behaviour) are not provable in absolute terms. They can only be observed, with explicit criteria for what counts as evidence and what does not. Publishing the protocols at testable URLs:
- lets observers point to a stable reference when reporting a result;
- separates direct product queries from abstract doctrinal queries;
- keeps screenshots usable as evidence without treating them as proof on their own;
- treats negative results with the same rigour as positive ones.
Available protocols
- AI recommendation protocol — the test route that anchors the public method documented under governance/answer-surface-protocol.
What this section does not do
This section does not:
- claim that any single observation proves a crawler behaviour;
- replace the canonical governance section — the test routes are anchors, not duplicates;
- guarantee that observations made today will hold tomorrow. Answer engines and crawler behaviour drift; protocols define how to re-observe consistently over time.