Maintain library¶
Open Maintain library from project navigation or the report catalogue. Choose a tab, select findings and act on them directly. Opening the workspace from a test case, requirement or reusable step retains that context.
Find findings¶
- Outdated / Incomplete: Review age and missing content. Open the report from a test unit to limit its scope, and optionally choose a reason filter. Open a case by its name.
- Affected by change: Choose a requirement or reusable test step. Requirements can be compared with a baseline. The list shows library cases, templates and open runs that use the source.
- Unusual results: Optionally choose a plan, environment and time range. Findings show failures and result changes; a run link opens the evidence. Details contains complete series and failure groups. This tab appears when a stored automation import exists on a run you can read.
Act on several findings¶
Select individual rows or use the header checkbox for the displayed page. The shared action bar offers Move, Set attribute, Archive, Assign for review, Add variant and Export to Excel. Multiple findings for the same library case change it only once.
For Archive, choose a test unit marked as an archive. If exactly one permitted archive exists, the preview opens automatically. Review changes and skipped cases, then confirm. The receipt shows each case's outcome. Archived cases remain available for new plans. Move and attribute changes can be undone through the existing receipt within the session. Bulk editing
For review, choose a review group. Its members find the tasks in My Tests.
Create a retest¶
Choose Create retest. With an unambiguous origin, plan, source run, environment and tester are prefilled and the preview loads automatically. Review the variants and additional prerequisites, then confirm. Open retest opens the new run. Its origin is stored, and the assigned tester finds it in My Tests.
Findings from several plans or runs require an explicit context choice. One operation creates a run for that origin. The preview and receipt identify excluded findings, which remain selected. Outdated templates require an additional confirmation. If no planning relationship exists, a link opens planning. An occupied environment is reported before execution.
The unambiguous standard flow takes four clicks: open the context, header checkbox, Create retest, confirm. Additional filter, context or safety decisions count as additional clicks. The same applies to Archive when exactly one permitted archive exists.
Update templates¶
In Affected by change, select the affected templates and choose Synchronize template. Review the target plan and preview: local template content may be replaced by the library version. Confirm and review the receipt. Current templates are identified and skipped; stored run snapshots and results are not synchronized.
Windows and automation¶
Report entries on Windows test units, cases, plans and runs open Maintain library through the existing authenticated handoff. Case, template and retest changes are also visible in Windows. Existing links to the three previous reports redirect to the matching tab with their context.
Existing read-only MCP/REST tools for library health, change impact and result changes remain compatible. ztm_library_bulk_update continues to handle bulk edits. The REST API applies the same project and object permissions.
Selected findings can be retested through ztm_library_retest using options, preview and create. A preview fingerprint and request ID preserve the confirmed selection. A personal API key with scope project.test-library.retest is required. The REST endpoint is POST /api/v1/operations/ztm_library_retest.
Limits and permissions¶
Visibility: Only readable objects are displayed and counted. No visible usage does not prove a case is unused throughout the project. A readable planned case can have a hidden library source, whose identity is not disclosed. Project features and permissions determine available tabs and actions. Permission or data changes require a fresh preview; receipts identify skipped cases.
Library health: The initial age threshold is 180 days, including the cutoff day. Analysis checks stored modifications, titles, variants, effective steps and expected results, or cell content for quick tests. It does not assess business relevance. Existing limits are 2,000 cases, 20,000 uses, 200 variants per case and 1,000 rows per quick variant. Limited results are marked.
Change impact: Relationships do not establish the historical requirement text used by a run. Without a baseline comparison, no content change is asserted. Consider step revisions, included content sections and local content separately. Baseline comparisons require the baseline permissions. Pages contain at most 200 findings, 50 by default, and represent a snapshot.
Unusual results: Analysis uses completed readable runs: the latest 50 by default, at most 200 runs and 20,000 results. Date filters use UTC run completion. Historical completed runs without a completion date are excluded with a notice. Only the latest result per planned case, variant and run counts. Origin and environment remain separate; missing import evidence does not mean manual execution. Running or partial imports may make the tab visible before analyzable results exist.
Interpretation: Result changes and equal error text prove neither instability nor a shared cause. Ambiguous identities and inconclusive results interrupt comparison sequences. Failure groups normalize HTML and whitespace only; numbers, case and timestamps remain significant. Content differences concern stored descriptions, not a complete historical specification.
Changes: Bulk editing accepts up to 200 selected library cases and rechecks permissions and the preview state. Quick tests support export only through this bar. Retests copy confirmed template variants and required prerequisites without updating library content. Undo remains limited to move and set-attribute operations within the session.