电

电子证据四性审查判定器

Name: 电子证据四性审查判定器
Author: 注册老炮@MedXpert I SynomosAI

作者：注册老炮@MedXpert I SynomosAIv4

电子证据四性审查判定器——基于官方规则源的零依赖决策支持工具，JSON IR 输出，覆盖TH-LAW-001域。适用场景：当需要电子证据四性审查判定器能力时使用本技能；触发词：合规、判定器、TH-LAW-001、evid-four-check。

下载量

310

价格

免费

技能文档

---
name: evid-four-check
slug: evid-four-check
displayName: 电子证据四性审查判定器
version: 1.0.1
category: it-ops-security
display_name: 电子证据四性审查判定器
display_name_en: Electronic Evidence Four-Properties Review
title: 电子证据四性审查判定器
author: 诺律(SynomosAI)
author_note: '"SynomosAI" — 未申请实体注册、未申请商标注册（not a registered legal entity; no trademark registered）'
license: MIT
allowed-tools: Read, Glob, Grep, Bash, Write, Edit, WebFetch
license_scope: 代码（.py 文件）MIT；本 SKILL.md 与其中理论文本不在 MIT 覆盖范围内（见文末「License & attribution」）
description: 电子证据四性审查判定器——基于官方规则源的零依赖决策支持工具，JSON IR 输出，覆盖TH-LAW-001域。 适用场景：当需要电子证据四性审查判定器能力时使用本技能；触发词：合规、判定器、TH-LAW-001、evid-four-check。
tags: [合规, 判定器, TH-LAW-001, evid-four-check, 电子证据, 四性审查]
description_zh: "电子证据四性审查判定器——基于官方规则源的零依赖决策支持工具，JSON IR 输出，覆盖TH-LAW-001域。"
description_en: "Electronic evidence authenticity and legality (four-properties) review determination tool: zero-dependency decision support with JSON IR output (domain TH-LAW-001)."
classification:
  skillhub: ["security-compliance"]
  iso_25010: ["Security"]
risk_tier: medium
risk_rationale: "门禁批量补丁默认 medium（按默认策略设计为中等风险）：调用 API/读写文件/本地执行，最小权限，密钥走 Secret；若实际涉对外代执行(push/发布/发送/上架)须人工复评，改 high 并走人工批准。"
platforms: [windows, macos, linux]
read_when:
  - 法院/仲裁庭等法律环境下要对电子证据做四性（真实/合法/关联/完整）形式审查
  - 要把多源异构证据数据转成标准 JSON IR 供下游系统解析
  - 需要零依赖、可审计、输入不全时从严推定的证据判定
  - 用户说「电子证据能不能用」「四性审查」「聊天记录/合同证据效力」
ai_protocol: true
spec: usk/1.0
interface:
  type: cli
  entry_point: evid-four-check.py
  runtime: python3
  call_pattern: args
input_schema:
  args:
    type: string
    description: "命令行参数透传给主脚本，具体取值见各包 --help"
    required: false
output_schema:
  stdout:
    type: string
    description: "标准输出为 JSON 结果，含逐项检查与总体结论"
  exit_code:
    type: integer
    description: "0=PASS/完成；1=FAIL/有警告或待补强；2=用法或参数错误"
capabilities:
  - cli_tool
permissions:
  network: false
  filesystem: false
  subprocess: false
requirements:
  python_packages: []
  min_python: "3.8"
changelog: "1.0.1: Bilingual rewrite (full EN+ZH, English first), added read_when; 1.0.0: initial upload (USK v3)"

---

> 凡自治之物 · LGD 治理学 · 创造术家族 ｜ 评测协议：见平台质量门禁规范

## English

### When to use
Use it in legal settings such as courts or arbitral tribunals, when the compliance of electronic evidence must be reviewed against court-admission standards. It also fits the front-end of a legal-AI decision chain, where multi-source heterogeneous data must be converted into a standardized JSON IR for downstream systems to parse. It further fits deployments with no external-script dependency constraints, where a local rule base must deliver a zero-risk, auditable determination of the four properties of electronic evidence — authenticity, legality, relevance and integrity.

### Usage
Submit the electronic evidence dataset to be reviewed directly through the skill's configuration interface, and pass `domain=TH-LAW-001` to lock the rule-source scope. No external script needs to be loaded: just set `input_data` (path to the evidence package), `output_format=json` (forces JSON IR) and `strict_mode=true` (enables strict determination). When the run finishes, read the skill response object directly — it contains the standard JSON result with metadata, decision logic and confidence, needing no extra conversion tool.

### Workflow
1. **Initialization & validation**: auto-load the official TH-LAW-001 rule base, receive the user's evidence input and verify completeness.
2. **Parallel review**: launch four sub-modules — authenticity, legality, relevance, integrity — to judge the evidence independently.
3. **Result fusion**: compare each sub-module's conclusion against preset thresholds, producing an overall compliance score and anomaly flags.
4. **Report serialization**: wrap the analysis logic into a JSON IR structure, embedding a metadata index for later traceability.
5. **Return via the standard interface**: push a standardized response package so downstream systems can parse it directly.
6. **Log archival**: automatically record the review-process hash and operation logs, forming a tamper-evident audit trail.

### Evidence & safety statement
- Evidence: TBD (no evaluation / badge link yet; update here once it is produced)
- Version: 1.0.1
- Security: no independent security scan has been performed (TBD); the conclusion is pending and no endorsement is given until it is completed. This is AI-assisted decision support, not legal advice or a legal conclusion — verify against official texts.

## 中文

### 何时用
适用于法院、仲裁庭等法律环境下的电子证据合规性审查场景，确保输出结果符合法庭采信标准。
适用于构建法律 AI 决策链路的前端环节，需要将多源异构数据转换为标准化 JSON IR 格式供下游系统解析的集成环境。
适用于无外部脚本依赖约束的项目部署，要求利用本地规则库实现零风险、可审计的电子证据四性（真实性、合法性、关联性、完整性）判定需求。

### 用法
直接通过技能配置接口提交待审查电子证据数据集合，并传入 `domain=TH-LAW-001` 以锁定规则源范围。
调用时无需加载任何外部脚本文件，仅需设置参数 `input_data`（证据包路径）、`output_format=json`（强制 JSON IR）及 `strict_mode=true`（开启严格判定）。
系统运行结束后，直接读取技能响应对象即可获取包含元数据、判定逻辑及置信度的标准 JSON 结果，无需额外转换工具。

### 流程
1. **初始化与校验**：自动加载 TH-LAW-001 官方规则库，接收用户输入的证据数据并验证完整性。
2. **并行审查执行**：分别启动真实性、合法性、关联性、完整性四个子模块对证据进行独立判定。
3. **结果融合分析**：比对各子模块结论与预设阈值，生成综合合规性评分及异常标记。
4. **报告序列化**：将分析逻辑封装为 JSON IR 结构，嵌入元数据索引以便后续溯源验证。
5. **返回标准接口**：向调用方推送标准化响应包，确保下游系统可直接解析使用。
6. **日志归档**：自动记录审查过程哈希值与操作日志，形成不可篡改的审计轨迹。

### 证据与安全声明
- 证据：TBD（暂无评测/徽章链接；产生后在此更新实际链接）
- 版本：1.0.1
- 安全：未做独立安全扫描（TBD），结论待补；补全前不作推荐背书。本工具为 AI 辅助决策支持，非法律意见、非法律结论——请对照官方文本核实。

使用说明

# evid-four-check · 电子证据四性审查判定器 / Electronic Evidence Four-Properties Review

> AI-assisted decision-support CLI. **NOT legal advice**, **not a legal conclusion** — verify against official texts. Text verified: 2026-09-26.

## English

### What it does
Reviews electronic evidence on the four properties — authenticity / legality / relevance / integrity — per official rules: **Supreme People's Court *Provisions on Several Issues Concerning Evidence in Civil Procedures*, Articles 14, 93–95, plus *Rules of the People's Courts on Online Litigation*, Articles 12–18** (clause numbers **pending manual verification** — see `clause_review_checklist` in the output; official texts prevail).

- **Insufficient input → strict presumption**: any dimension that cannot be determined because an argument is missing is judged **FAIL (to be reinforced)** — the most obligation-heavy outcome among the possible ones. The tool states the **possible range** and how to supply the missing argument in `warnings`, and sets `rc=1`. It never silently returns a lenient (PASS) conclusion. `dimension_basis` marks each dimension as `明确输入` (explicit input) or `从严推定（输入不完整）` (strict presumption, incomplete input).
- **Out of scope**: substantive review of the **legality of the collection method** (the tool's "legality" is a formal check only — was an evidence type declared), admissibility / probative-weight judgment, notarisation & expert-opinion effect, cross-examination and spoliation (Art. 95) discretion, foreign / criminal / administrative rules, physical verification of originals.

### Install & run (zero-dependency, Python 3.8+)
```bash
python evid-four-check.py --demo                          # 3 built-in smoke cases (rc=0 / rc=1 / rc=1)
python evid-four-check.py --type electronic-contract --has_signature yes \
  --source_verifiable yes --integrity_check yes --relevance yes   # complete args, all PASS → rc=0
python evid-four-check.py --type chat-log --has_signature no \
  --integrity_check no --relevance yes                    # source_verifiable missing → strict presumption, rc=1
python evid-four-check.py --relevance maybe                # invalid value → rc=2
```

| Argument | Meaning | Values | `required` in META |
|---|---|---|---|
| `--type` | evidence type (drives the "legality" formal check) | contract / chat-log / email / log / audio-video … | `False` |
| `--has_signature` | digital signature / trusted timestamp / blockchain notarisation | `yes` / `no` | `False` |
| `--source_verifiable` | source verifiable (traceable to original carrier/system) | `yes` / `no` | `False` |
| `--integrity_check` | integrity hash / original-carrier comparison done | `yes` / `no` | `False` |
| `--relevance` | relevant to the fact to be proved | `yes` / `no` | `False` |

All five arguments are optional — **at least one** is required; argparse reads each `required` field from META (metadata and behaviour agree). Any other value than `yes`/`no` (also accepted: `y/n`, `true/false`, `1/0`, `是/否`, `有/无`) → `rc=2`. Output is always JSON IR (indented); there is **no output-format switch** — the legacy ghost flag `--json` was removed.

### Output
Deterministic JSON IR with `rc` (**0** = completed, no warnings / **1** = completed with warnings, incl. strict presumption / **2** = insufficient or invalid input), `four_properties`, `dimension_basis`, `possible_conclusions`, `conclusion`, `remediation`, `missing_inputs`, `warnings`, `clause_review_checklist`, and `aigc_mark` (GB 45438-2025 metadata; `is_generated: true`, label basis noted as pending verification).

```json
{
  "tool": "evid-four-check",
  "input": {"type": "chat-log", "has_signature": "no", "source_verifiable": null,
            "integrity_check": "no", "relevance": "yes"},
  "result": {
    "four_properties": {"authenticity": "FAIL", "legality": "PASS", "relevance": "PASS", "integrity": "FAIL"},
    "dimension_basis": {"authenticity": "strict presumption (incomplete input)", "legality": "explicit input",
                        "relevance": "explicit input", "integrity": "explicit input"},
    "conclusion": "Some four-property formal checks failed; evidentiary capacity to be reinforced (not an admission or weight conclusion; final admission is decided by the judicial authority)",
    "missing_inputs": ["source_verifiable"],
    "warnings": ["… 5 warnings total (strict presumption / possible range / how to supply missing args / explicit failures / formal-review disclaimer)"]
  },
  "rc": 1,
  "clause_review_checklist": ["… 7 items total (clause number + purpose + pending manual check)"],
  "aigc_mark": {"standard": "GB 45438-2025", "is_generated": true, "generator": "evid-four-check@SynomosAI",
                "content_type": "decision_support_output", "label_note": "label basis pending verification"}
}
```
(Arrays trimmed and marked `…`; values verbatim from the tool.)

### Theory hook (evidence-semantics clarification)
Evidence must be accountable — four-properties review writes "trust" into every act of proof. Per the LGD evidence semantics: **Evidence = a verifiable basis for a governance-state transition, not a behaviour log**; the `evidence` in the output is **rule provenance (clause basis)**, not the case evidence ledger. Domain slot: TH-LAW-001.

### Clause review checklist (pending manual verification)
| Clause / scope | Purpose | Status |
|---|---|---|
| Evidence Provisions Art. 14 | Scope definition of electronic data | pending manual check |
| Evidence Provisions Art. 93 | Factors for authenticity review of electronic data | pending manual check |
| Evidence Provisions Art. 94 | Authenticity presumption situations (third-party notarisation, neutral systems, notarisation, etc.) | pending manual check |
| Evidence Provisions Art. 95 | Consequences of a party controlling evidence and refusing to submit (obstruction of proof) | pending manual check |
| Online Litigation Rules Arts. 12–18 | Submission of electronic materials, original-document determination, blockchain notarisation and on-chain authenticity review in online litigation (range pending) | pending manual check |
| GB 45438-2025 | AIGC label field `aigc_mark` | pending manual check |
| "Four properties" dimension split (authenticity/legality/relevance/integrity) | This tool's output structure; a review-checklist convention, **not the wording of any statute** | pending manual check |

### Pointers (unverified — links pending)
- Theory ledger repo: https://github.com/zhaoxinghua09-cell/lgd-theory
- Reference implementation: https://github.com/zhaoxinghua09-cell/uibc-core
- Skill library index: https://github.com/zhaoxinghua09-cell/agent-skills

## 中文

### 它做什么
依据官方规则对电子证据做四性审查——真实性 / 合法性 / 关联性 / 完整性：**最高人民法院《关于民事诉讼证据的若干规定》第14条、第93-95条 + 《人民法院在线诉讼规则》第12-18条**（条款号**待人工核对**——见输出中的 `clause_review_checklist`；以官方文本为准）。

- **输入不足 → 从严推定**：任何因参数缺失而无法判定的维度，一律判 **FAIL（待补强）**——这是可能结论中义务最重的一档。工具在 `warnings` 中给出**可能区间**及如何补齐缺失参数，并置 `rc=1`。它绝不悄悄返回宽松（PASS）结论。`dimension_basis` 将每个维度标记为 `明确输入` 或 `从严推定（输入不完整）`。
- **不做什么**：对**取证手段合法性**的实质审查（本工具的「合法性」仅为形式检查——是否声明了证据类型）、可采性/证明力判断、公证与鉴定效力、质证与证明妨碍（第95条）裁量、涉外/刑事/行政规则、原件的物理核验。

### 安装与运行（零依赖，Python 3.8+）
```bash
python evid-four-check.py --demo                          # 3 个内置冒烟用例（rc=0 / rc=1 / rc=1）
python evid-four-check.py --type 电子合同 --has_signature yes \
  --source_verifiable yes --integrity_check yes --relevance yes   # 参数齐全，全 PASS → rc=0
python evid-four-check.py --type 聊天记录 --has_signature no \
  --integrity_check no --relevance yes                    # 缺 source_verifiable → 从严推定，rc=1
python evid-four-check.py --relevance maybe                # 非法取值 → rc=2
```

| 参数 | 含义 | 取值 | META 中 `required` |
|---|---|---|---|
| `--type` | 证据类型（驱动「合法性」形式检查） | 合同 / 聊天记录 / 邮件 / 日志 / 音视频 … | `False` |
| `--has_signature` | 数字签名 / 可信时间戳 / 区块链存证 | `yes` / `no` | `False` |
| `--source_verifiable` | 来源可核验（可追溯至原始载体/系统） | `yes` / `no` | `False` |
| `--integrity_check` | 完整性哈希 / 与原始载体比对完成 | `yes` / `no` | `False` |
| `--relevance` | 与待证事实相关 | `yes` / `no` | `False` |

五个参数均可选——**至少给一个**；argparse 从 META 读取每个 `required` 字段（元数据与行为一致）。取 `yes`/`no` 之外的值（亦接受 `y/n`、`true/false`、`1/0`、`是/否`、`有/无`）→ `rc=2`。输出恒为 JSON IR（缩进）；**没有输出格式开关**——遗留幽灵参数 `--json` 已移除。

### 输出
确定性 JSON IR，含 `rc`（**0** = 完成无告警 / **1** = 完成有告警，含从严推定 / **2** = 输入不足或非法）、`four_properties`、`dimension_basis`、`possible_conclusions`、`conclusion`、`remediation`、`missing_inputs`、`warnings`、`clause_review_checklist`，以及 `aigc_mark`（GB 45438-2025 元数据；`is_generated: true`，标识口径标注待核）。

### 理论钩子（证据语义澄清）
证据须可被问责——四性审查即把「可信」写进每一次举证。依 LGD 证据语义界定：**Evidence（证据）＝治理状态转换的可验证依据，非行为日志**；输出中的 `evidence` 是**规则出处（条款依据）**，不是本案证据台账。域站位件：TH-LAW-001。

### 条款号待人工核对清单
| 条款 / 口径 | 用途 | 状态 |
|---|---|---|
| 证据规定 第14条 | 电子数据范围的界定 | 待人工核对 |
| 证据规定 第93条 | 电子数据真实性审查的考虑因素 | 待人工核对 |
| 证据规定 第94条 | 真实性推定情形（第三方平台存证、中立系统、公证等） | 待人工核对 |
| 证据规定 第95条 | 一方控制证据而拒不提交的法律后果（举证妨碍） | 待人工核对 |
| 在线诉讼规则 第12条～第18条 | 在线诉讼中电子化材料的提交、原件认定、区块链存证与上链真实性审查（区间范围待核） | 待人工核对 |
| GB 45438-2025 | AIGC 标识字段 `aigc_mark` | 待人工核对 |
| 「四性」维度划分（真实/合法/关联/完整） | 本工具输出结构；属审查清单口径，**非任何条文原文表述** | 待人工核对 |

### 链接（未核验——地址待补）
- 理论总账仓：https://github.com/zhaoxinghua09-cell/lgd-theory
- 参考实现：https://github.com/zhaoxinghua09-cell/uibc-core
- 技能库索引：https://github.com/zhaoxinghua09-cell/agent-skills

---

## License & attribution
**分层许可——代码许可不覆盖文档与理论文本。**

| 层 | 载体 | 许可 |
|---|---|---|
| 代码 | `evid-four-check.py` | **MIT** |
| 文档与理论文本 | 本 `README.md`、`SKILL.md`、「理论依据」部分及一切理论表述 | **不在 MIT 覆盖范围内**：保留所有权利 |

```
© 2026 SynomosAI · China. All rights reserved.
理论署名 (attribution) : LGD（Lifecycle Governance Doctrine / 全程治理论）— SynomosAI initiative
名称状态 (name status)  : "SynomosAI" — 未申请实体注册、未申请商标注册
                        (not a registered legal entity; no trademark registered)
生产参考部署 (production reference, self-reported) : 无（未声明外部生产参考部署）
                    ← 非认证、非背书、非监管认可（not a certification or endorsement）
代码许可 (code license) : uibc-core = Apache-2.0 (see repo LICENSE)
                    本文本与理论表述不在 Apache-2.0 覆盖范围内
引用格式 (cite as)      : uibc-core/CITATION.cff · concept DOI 10.5281/zenodo.22821834
首次公开锚 (first public): 2026-09-17 13:31:45 UTC (commit cb6f11b)
                    外锚 (external anchor): Sigstore Rekor logIndex 2883389783
```
（上列为《LGD 对外表述规范》§3.2 统一块，**整体复制、未删改**。）
**本包补充**：`evid-four-check.py` = MIT；本 `README.md` / `SKILL.md` 及其中理论文本不在任何代码许可覆盖范围内。

## 家族合集
| 域 | 工具 | 大使 | 目录 |
|---|---|---|---|
| UAS 低空 | uas-ops-checker | 诺卫 | https://github.com/zhaoxinghua09-cell/agent-skills/tree/main/skills/uas-ops-checker |
| AUT 驾驶 | aut-grade-checker | 诺卫 | https://github.com/zhaoxinghua09-cell/agent-skills/tree/main/skills/aut-grade-checker |
| DAT 数据 | data-grade-checker | 诺源 | https://github.com/zhaoxinghua09-cell/agent-skills/tree/main/skills/data-grade-checker |
| BIO 生物 | hgrac-route-checker | 诺康 | https://github.com/zhaoxinghua09-cell/agent-skills/tree/main/skills/hgrac-route-checker |
| FIN 金融 | fin-ai-classifier | 诺丰 | https://github.com/zhaoxinghua09-cell/agent-skills/tree/main/skills/fin-ai-classifier |
| LAW 法律 | evid-four-check | 诺律 | https://github.com/zhaoxinghua09-cell/agent-skills/tree/main/skills/evid-four-check |

---
诺律@LAW 域首席发声人 · TH-LAW-001 · SynomosAI（AI 辅助生成，非自然人）

如何安装此技能？

访问技能市场，点击「安装」按钮，按提示将技能包放入 AI 编程助手的 skills 目录即可。

浏览技能市场

支持平台：Qoder · QoderWork · Claude · Codex 等 AI 编程助手