基于证据的事实核查、来源可信度评估的Codex技能， 错误信息风险评分、预先揭露简报和HTML事实核查卡。

本版本保持SKILL.md简短，并采用渐进式披露：

• SKILL.md 包含触发元数据、护栏、模式选择和核心 工作流程。
• references/ 包含详细的源代码评估、红旗、MFS， 保加利亚/欧盟来源，保加利亚信息空间背景，媒体素养 （教育建议，包括保加利亚本土基础），以及输出指导。
• schema/ 定义了 JSON 结果契约。
• scripts/ 验证结果，渲染 HTML 卡片，并构建干净的上传 档案馆。
• tests/ 包含标准、比较和预掩的回归调剂 模式。

验证

根据仓库根：

渲染样本卡

建立档案

以fact-check/为根的典籍风格档案：

ChatGPT 在归档根带 SKILL.md 上传归档：

生成的归档会有意排除测试、out/、dist/、缓存文件和 之前的包伪影。

本地法典安装

通过复制 fact-check 目录安装到：

Fact Check Skill

Codex skill for evidence-led fact-checking, source credibility evaluation, misinformation-risk scoring, prebunking briefings, and HTML Fact-Check Cards.

This version keeps SKILL.md short and uses progressive disclosure:

• SKILL.md contains trigger metadata, guardrails, mode selection, and the core workflow.
• references/ contains detailed source-evaluation, red-flag, MFS, Bulgarian/EU source, Bulgarian information-space context, media-literacy (educational tips, incl. a native Bulgarian base), and output guidance.
• schema/ defines the JSON result contract.
• scripts/ validates results, renders HTML cards, and builds clean upload archives.
• tests/ contains regression fixtures for standard, comparison, and prebunking modes.

Validate

From the repository root:

python .\scripts\validate_skill.py .
python -m unittest discover -s tests

Render a Sample Card

python scripts\render_card.py tests\fixtures\standard_health.json --output out\standard_health.html

Build Archives

Codex-style archive with fact-check/ as the archive root:

python scripts\package_skill.py --layout folder --output dist\fact-check-codex.zip

ChatGPT upload archive with SKILL.md at the archive root:

python scripts\package_skill.py --layout root --output dist\fact-check-chatgpt-upload.zip

Generated archives intentionally exclude tests, out/, dist/, cache files, and previous package artifacts.

Local Codex Install

Install by copying the fact-check directory to:

C:\Users\******\.codex\skills\fact-check

Keep a backup of any existing installed skill before replacing it.