Precision and recall measure a classifier or retriever from two angles. Precision = of the items you flagged positive, what fraction were actually positive. Recall = of all the actually-positive items, what fraction you caught. They trade off against each other.
Worked example: a spam filter that flags only the 3 most obvious spams has high precision (all 3 right) but low recall (missed 20 others); flagging everything has perfect recall but terrible precision. The F1 score is their harmonic mean, used when you want one number. Gotcha: which matters more depends on cost — recall for cancer screening (a missed case is catastrophic), precision for a legal-hold search (false positives are expensive to review).