I’m interested in understanding how people can researchReference Materials
Replay
Every recorded step of a run, as the agent saw the screen: the frame, the action it took, what it said, and which rubrics the judge's evidence had reached by then. 4 runs recorded with screenshots.
Claude Opus 5.5
- 100%
- Rubric score
- 100
- Steps
- 6/6
- Rubrics passed
Task promptI’m interested in understanding how people can research what a public website actually said or displayed in the past, rather than assuming that the current version of a webpage represents its historical content. I want to compare several publicly accessible web-archiving and preservation approaches and investigate what can be verified about archived pages, capture dates, missing content, redirects, images, downloadable documents, embedded resources, and changes between versions. I’m particularly interested in how a researcher can distinguish an archived copy from a current page, identify whether an archived capture is complete, and deal with situations where a page exists in search results but its historical content cannot be recovered. I also want to understand the limitations of web archives, including robots exclusions, broken assets, dynamic pages, changing URLs, and incomplete capture histories. Rather than simply listing web-archive services, I want to reconstruct a small set of historical webpage changes and document the evidence trail used to establish what changed. I’d also like to hear from researchers, journalists, archivists, and web users about practical difficulties they encounter when trying to recover older online information. Please research this in a real browser and create a practical web-history verification guide that another researcher could reuse. Use publicly accessible pages that can be read without logging in, subscribing, or paying. Prefer recognised web-archiving organisations, libraries, universities, government digital-preservation programmes, archival institutions, and established research organisations for factual information. Use archived pages and original websites as evidence where accessible, and clearly distinguish archival records from later commentary. Use community discussions only for practical experiences and clearly separate them from archival evidence. If a selected source is inaccessible, use another qualifying public source and explain the replacement. If a requested historical detail cannot be established from accessible evidence, record it as "not verified." Do not invent historical webpage content, capture dates, missing-page explanations, or claims about why a page disappeared.
I’m interested in understanding how people can research what a public website actually said or displayed in the past, rather than assuming that the current version of a webpage represents its historical content. I want to compare several publicly accessible web-archiving and preservation approaches and investigate what can be verified about archived pages, capture dates, missing content, redirects, images, downloadable documents, embedded resources, and changes between versions. I’m particularly interested in how a researcher can distinguish an archived copy from a current page, identify whether an archived capture is complete, and deal with situations where a page exists in search results but its historical content cannot be recovered. I also want to understand the limitations of web archives, including robots exclusions, broken assets, dynamic pages, changing URLs, and incomplete capture histories. Rather than simply listing web-archive services, I want to reconstruct a small set of historical webpage changes and document the evidence trail used to establish what changed. I’d also like to hear from researchers, journalists, archivists, and web users about practical difficulties they encounter when trying to recover older online information. Please research this in a real browser and create a practical web-history verification guide that another researcher could reuse. Use publicly accessible pages that can be read without logging in, subscribing, or paying. Prefer recognised web-archiving organisations, libraries, universities, government digital-preservation programmes, archival institutions, and established research organisations for factual information. Use archived pages and original websites as evidence where accessible, and clearly distinguish archival records from later commentary. Use community discussions only for practical experiences and clearly separate them from archival evidence. If a selected source is inaccessible, use another qualifying public source and explain the replacement. If a requested historical detail cannot be established from accessible evidence, record it as "not verified." Do not invent historical webpage content, capture dates, missing-page explanations, or claims about why a page disappeared.