cff-version: 1.2.0
message: "If you use this software or its measurements, cite it as below."
title: "HiveBench"
abstract: >-
  Evaluation suite and Studio sidecar for Splinter-Memory. Ships the white
  paper's falsifiable predictions (P1-P11) as executable, offline tests with
  measured verdicts, deterministic retrieval diagnostics that avoid
  LLM-as-judge circularity in the evidence path, and replayable paired A/B
  batteries (splinter-curated context vs a naive FIFO window).
type: software
authors:
  - name: "sky-is-green"
    alias: "sky-is-green"
version: "0.1.0"
date-released: "2026-09-22"
license: MIT
repository-code: "https://github.com/sky-is-green/hivebench"
url: "https://github.com/sky-is-green/hivebench"
keywords:
  - evaluation
  - benchmark
  - context-management
  - llm
  - reproducibility