Make benchmark identity explicit in metadata

- pretty_name: 'CAP-Bench: Cross-Site Browser-Agent Benchmark' so the
page header reads as a benchmark, not a generic dataset.
- Add task_ids: open-domain-qa, annotations_creators / language_creators
/ source_datasets / multilinguality fields so HF's metadata schema
treats the repo as a fully-specified evaluation dataset.
- Reorder tags so 'benchmark' / 'agent-evaluation' lead. Add
'llm-evaluation' and 'agent-as-a-judge' tags so the repo surfaces
under benchmark-related searches.

README.md

@@ -5,17 +5,30 @@ language:
 task_categories:
 - text-generation
 - question-answering
-pretty_name: CAP-Bench
+- other
+task_ids:
+- open-domain-qa
+pretty_name: 'CAP-Bench: Cross-Site Browser-Agent Benchmark'
 size_categories:
 - n<1K
+annotations_creators:
+- expert-generated
+language_creators:
+- expert-generated
+source_datasets:
+- original
+multilinguality:
+- monolingual
 tags:
+- benchmark
+- agent-evaluation
 - agents
 - browser-agents
 - web-agents
-- benchmark
-- agent-evaluation
 - cross-site
 - gui
+- llm-evaluation
+- agent-as-a-judge
 configs:
 - config_name: default
   data_files: