Editorial and Tool-Testing Standards — OpusDesk Hub Tools
Practical, checkable guidance
Guides should describe the actual task, supported inputs, a reproducible example and practical limitations. Original utility guides are adapted with current routes and branding while their examples and cautions are retained. A published page is not evidence that every browser or maximum input has been tested.
Verification scope
A tool check should cover its primary workflow, independent expected results, empty or invalid input and the relevant copy or download action. Browser checks also examine keyboard access, narrow screens and errors. Reports must identify what was actually run; a build or screenshot alone does not prove an operation works.
Privacy claims
Local processing concerns the input and result, not every request made while loading the site. Source inspection and browser network checks must distinguish asset downloads from a file upload. This build removes the original advertising and analytics integrations; do not infer that linked external services share the same behaviour.
Limits and responsible interpretation
Device memory, browser support and input conventions matter. JWT decoding is not signature verification. English grammar checks use heuristics. Text similarity is not a web-wide plagiarism search. An image-based PDF is not OCR. A grade calculation is only as valid as the supplied weights and institutional rules.
Corrections and dates
Dates retained in guides describe the source material and do not certify a new deployment or recent browser test. Report a problem through the contact page with a safe example, expected output and browser details. Corrections should update both instructions and behaviour when needed.
Independence
No advertising or analytics integration is enabled in this build. Guide recommendations are based on the task and documented limitations rather than invented usage rankings, reviews or performance claims.