← Week 6 portalMaster file in the engagement folder:
my-work/week-6/fund-level/library/applicability/applicability-screen-v1.md

Applicability screen (v1, scripted)

Record ids. D-6-n is a Week 6 decision, J-n a journal entry, F-0n a review finding, DR6-n a data request, A-n an assumption; all are in the engagement record.

Rules: applicability-rules-v2.csv has 25 rules, one per framework, entity type (fund, GP, company or any) and route. Each rule has a machine-readable condition, the route (mandatory by law, requested by LPs, requested by customers, requested by a buyer in diligence, or none), the deciding threshold, a source URL, a retrieval date, a verification label and a re-check-by date. The v1 rules, written as prose, are in _archive/.

Input: entity-facts-v1.json in the case folder. It holds the facts for each entity. The allowed fact names and types are listed in engine/run_applicability.py (FACTS): revenue, EU group employees, AUM, California business, EU, UK and ISSB-jurisdiction listing, FCA authorisation, PRI signatory, EDCI participation, exit in view, and the lists of LP requests, customer requests and what the GP collects.

Fact definitions that carry legal tests:

Run: python3 outputs/engine/run_applicability.py --case <case folder> (also part of run-all.sh). Output goes to the case's out/:

How it decides

Lens: mandatory routes are not negotiable. LP, customer and buyer routes are scoped with the owner on whether serving the request reduces deal risk or supports exit value.

Independent check: checks/chk_applicability.py (applicability-sound). It recomputes the California and FCA outcomes from the facts using thresholds declared in the check. It also confirms that every entity met every rule for its type exactly once, and that every undecided rule is surfaced.

Thresholds marked unverified must be checked against the source before they decide anything for a client.

Portfolio edition: Kith Climate cohort case, fictional client.