Datasets:
Download data/schema.json from AriaAICompany/web-watch-lab: direct link, hf CLI and curl.
- Browser
- Download file 517 Bytes
-
https://huggingface.co/datasets/AriaAICompany/web-watch-lab/resolve/main/data/schema.json
- Command line
-
hf download hf://datasets/AriaAICompany/web-watch-lab/data/schema.json
-
curl -L -o schema.json https://huggingface.co/datasets/AriaAICompany/web-watch-lab/resolve/main/data/schema.json
517 Bytes
| { | |
| "schema_id": "webwatch.request_window.v1", | |
| "request_fields": [ | |
| "request_id", | |
| "timestamp", | |
| "site_id", | |
| "method", | |
| "path_norm", | |
| "query_redacted", | |
| "status", | |
| "latency_ms", | |
| "bytes", | |
| "client_alias", | |
| "user_agent_class", | |
| "deploy_marker" | |
| ], | |
| "timezone": "UTC", | |
| "null_meaning": "dropped as incomplete", | |
| "window_minutes": 5, | |
| "notes": "Query strings, tokens, and emails are redacted before scoring. Client values are aliases, not production IPs." | |
| } |