judgment

基于法律裁判文书的事件抽取及其应用，包括数据的分词、词性标注、命名实体识别、事件要素抽取和判决结果预测等内容

Practitioner-judgment skill for Claude Code, Codex, and AI agents under ambiguity, trade-offs, and incomplete information.

JEAM is a Python package for evidence accumulation modeling of continuous judgments

Research preview for a portable, vendor-neutral representation of organizational judgment.

⚖️ End-to-End Judicial Problem Solving Platform powered by a fine-tuned Legal Llama 3.2 (GGUF), statutory RAG over Indian Law (IPC, CrPC, IT Act, Constitution), document OCR, judgment summarizer, and a physics-driven GSAP monochrome interface.

代入真实历史决策点、与史实对账来训练判断力——每次互动闭成数据闭环，判断力可 持续改善。Claude Code 技能。

DEPRECATED: this clone-and-go sandbox is no longer maintained and will be replaced by a hosted application. The open local experience is judgment-pack-desk.

Experimental multi-role decision review CLI with model-based evidence critique and an inspectable Markdown ledger.

A framework and evaluation for the rigor-deficit account of agentic AI failure on judgment work. Pre-registered, blinded, dual-rater study (n=30) with reproducible scoring artifacts.

MCP server for searching Taiwan judicial judgments (司法院裁判書系統) — 2 AI-callable tools over Model Context Protocol

把 Jev（一个小而快的判定器）接进你的 AI 助手：一个多月的实战手册——接入 / 判定点 / 影子期评测 / 踩坑集

SEAL — No seal, no advance. Workflow kernel: product truth as a sealed graph (not an agent parade, not Jev-with-UI).

Three focused Codex skills for bounded judgment, correction, and recurring friction.

The Judgment Exchange: dialects, the declarative translations between a system and a judgment pack, in both directions, held as hand-maintained data with fixtures and verification records

Clean-room evaluator implementations generating RFC 0006's evidence. Experimental; claims no JPS conformance.