Evidence librarian / EXAMPLE DESIGN
Finding a document is only useful when it answers the right question. See how the Evidence Librarian puts the passage, version and period together so a practitioner can check its relevance.
Working browser demonstration · Fictional sources · Local rules, no live agent
One case question and the internal records that case is allowed to use.
A passage register with exact excerpts, source versions and applicability gaps.
A qualified practitioner decides whether each source applies. A record from the wrong period remains a gap to resolve.
Change the reviewer hours, then request an allocation from the wrong week. The register shows the mismatch without turning it into a finding. Set the input and select “Prepare this result”. Then read the output before recording a fictional review.
Changing an input starts a fresh example.
Case-scoped internal search; no public-web egress or cross-client corpus.
PREPARED OUTPUTA passage register with exact excerpts, source versions and applicability gaps.
HUMAN REVIEWA qualified practitioner decides whether each source applies. A record from the wrong period remains a gap to resolve.
Retrieve the source, version and exact passage behind a weekly capacity question. A record from the wrong week stays visible as a conflict.
Retrieve the source, version and exact passage behind a weekly capacity question. A record from the wrong week stays visible as a conflict.
Local fictional example · 0 recorded events · Revision 1
These fictional originals belong to the selected case. Compare the output with the source, version and allowed use. Showing a source here does not clear missing evidence or grant wider access.
Synthetic source. Case D-01 asks for a one-week role-capacity comparison, not a staffing recommendation or revenue forecast. The target is simulation week 1. There are 6 comparable candidate packs of qualified demand, 40 analyst hours and 8 partner hours. R is available qualified-reviewer hours, a whole number from 4 to 24, default 12.
Use boundary: Synthetic fixture; private review within this fictional case only.
Synthetic source. For simulation week 1 only: analyst availability 40 hours; qualified-reviewer availability R hours; partner availability 8 hours. R comes from the selected fictional allocation scenario. Allocation describes available active effort, not signed client acceptance or bankable revenue.
Use boundary: Synthetic fixture; private review within this fictional case only.
Synthetic source. The reviewer allocation of R hours is labelled simulation week 0. No week-1 reviewer allocation is supplied in this exception fixture. Analyst and partner week-1 totals remain 40 and 8. Do not treat the old reviewer number as zero or as verified current capacity.
Use boundary: Comparison only; wrong-period evidence cannot support the target week's conclusion.
Synthetic source. Per comparable pack, analyst effort is 8 hours and partner effort is 2 hours under both methods. Current reviewer effort = 3 review + 1 correction = 4 hours. Proposed reviewer effort = 2 review + 0.5 correction + 0.5 operating = 3 hours. For each method, whole-pack capacity = min(floor(40/8), floor(R/reviewer hours per pack), floor(8/2), 6 demand). Floor each role before taking the minimum. No client acceptance, quality equivalence or realised gain is established by these assumptions.
Use boundary: Synthetic fixture; private review within this fictional case only.
Synthetic source. For D-01, retrieve the question, reviewer allocation and method for simulation week 1. R is the selected whole number from 4 to 24. Ordinary retrieval returns DEL-SCOPE v1, DEL-ALLOCATION v1 and DEL-METHOD v1. The challenge substitutes DEL-ALLOCATION-OLD v0, explicitly week 0, for the current allocation. The register must display that mismatch and request the week-1 record, without using a comparison specimen as a silently supplied correction. Analyst 40 hours and partner 8 hours remain supported for week 1. Report source passages and applicability, not a throughput finding. No other case or public-web search is included.
Use boundary: Fictional case metadata for this role example only.
Run the connected case to see what arrives before this step, who reviews its output and what happens when the next owner returns it.
Bring the task, its current owner and where it gets stuck to a no-fee Workflow Fit Call. If a paid System Blueprint is useful, it measures one service line and fully designs three connected workflows. Roles are chosen for those designs. Installation is separately scoped; pricing is quote-specific.
