SEMANTIC-CASE: 事实与陈述区分
- id: CASE-FACT-STMT
- 触发：输入含一方当事人陈述 + 一方客观证据相互矛盾（见 materials/familiar-sample.md M4/M5）
- 期望行为：候选须区分"客观证据"（合同、登记、函件）与"当事人陈述/争议性材料"，不得混同为已证事实。
- 判定要点：
- 产出中陈述类材料须标注为"陈述/待核实/证明力弱"；
- 核心事实结论须指向证据来源，不得把销售人员承诺直接当作事实。
- 来源：核心原则第 2、6 条。
/legal-skill-evaluation
@9be736d法律 Skill 分层质量评测工具。消费 skill-lint 的通用质量结论，再用三份测试材料、通用六维度、场景微调和律师 taste 评估法律产出，并定位最小修复单元。本技能应在审查、回归验证或发布验收法律 Skill 时使用。不要用于代替通用 Skill lint、正式法律意见或跨场景排名。
Ask your Agent
Use this Skill: https://skilld.dev/gh/cat-xierluo/legal-skills/legal-skill-evaluation
This session only. Nothing lands on disk.
≈139 tokens on demand. Your agent reads this file only when SKILL.md points to it.
Source: SKILL.md on GitHub
Third-party checks
No alerts10d3 checks · Risk SAFE
- Gen Agent Trust Hub10d
A comprehensive evaluation framework for legal AI skills designed to assess legal accuracy, fact faithfulness, and adherence to professional tastes. It includes Python scripts for automated quality gates, hashing, and execution probes. No malicious patterns were detected; the automated scripts are functionally aligned with the skill's primary purpose as a testing and evaluation harness.
- Socket10d
No alerts
- Snyk10d
Risk: LOW · No issues
Provenance
Signed by skilld at 9be736d. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.
Last checked against GitHub yesterday.
Activeupdated 2 weeks ago
Capability
- version
- 0.8.12
- author
- 杨卫薪律师（微信ywxlaw）
- homepage
- https://github.com/cat-xierluo/legal-skills