Public handling: selected full-text prompt with secret filtering.
Prompt Text
In repo D:\Research\TRUCE-Rec, audit the current baseline/fairness/provenance docs and configs. Focus on how to revise them to match this policy: all compared LLM baselines use Qwen3-8B base model, LoRA adaptation, official implementation/source where possible, official default/optimal hyperparameters for baselines, Ours may tune hyperparameters; shared TRUCE dataset/candidates/evaluator/score schema. Identify exact files/sections likely needing edits and any contradictions. Do not edit files. Return concise recommendations with file paths.
Reuse Notes
EXTRACTED: This is a selected Codex prompt or automation prompt from the local Codex corpus.
INFERRED: Future agents can reuse its structure, constraints, and acceptance criteria when creating similar Codex workflows.