F14

Context-store and task policy

The source summary is shown as authored. Follow its evidence links for study-specific status and cohort limits.

Source-reported narrativeValues in this source section that are not tied to a claim-ledger record remain source-reported and are not promoted to verified results.

Findings source section · FINDINGS.md

Hypothesis: H148–H153 ask whether context stores, task fairness checks, or harness mining improve retrieval or downstream Builds. Offline retrieval observations do not show that a novel context store improves Build success.

What ran: The public pages cover offline store retrieval and freshness, an incomplete context-preparation cohort, a forced-compaction pilot, an interim frontier lane, and a mixed-version harness inventory.

Result: Numerator/denominator: not recorded for a matched live Build comparison. Unit: offline retrieval, freshness, and inventory measures are separate from official Build passes. Date: no common result date recorded. Status: interim or source-reported offline diagnostic; no context-store efficacy result.

What it does not show: Retrieval ties, freshness checks, or mixed-version inventory do not establish better live task completion or identify a causal harness feature.

Evidence: F14 hypothesis page, context store, context cohort, E06 context, frontier, harness mining. Spec map: deferred scope and implementation freedom.