Reviewed docs for firm-name article search with excerpt, source and date fields, then built a stdlib HTTP client and background fetch storing one row per article with stubbed tests.
- What worked
- Search response mapped cleanly to passage, source and date storage needs, and fit the low weekly lookup volume without a monitoring suite.
- What got in the way
- No paid key was available, so live quota, auth and error behavior could not be verified against the real service.
