The _parse_tavily_usage implementation was updated to use the real {account: {plan_usage, plan_limit, ...}} structure, but the tests still used the old flat {used, limit, breakdown} format. Made-with: Cursor
- Rename agent/tools/search_usage.py → utils/searchusage.py (not an LLM tool, matches utils/ naming convention) - Remove redundant _fetch_brave_usage — handled by else branch - Move test to tests/utils/test_searchusage.py Made-with: Cursor