{
 "title": "ai.responsibleailabs/rail-score",
 "site": "台灣 AI 代理人任務中樞",
 "canonical": "https://penguindriver.com/s/ai.responsibleailabs%2Frail-score",
 "markdown": "https://penguindriver.com/s/ai.responsibleailabs%2Frail-score.md",
 "summary": "Responsible-AI guardrails for agents: scoring with policy, injection & PII detection, DPDP.",
 "date_modified": null,
 "retrieved": "2026-10-17",
 "content_sha256": "6ed66fc604bee437c203008a6861e85f25f237f20be444edee8b682393c1fa0f",
 "fingerprint_basis": "Markdown 版全文，移除贊助行與開頭 front matter，去除頭尾空白",
 "verify": {
  "text_snapshot": "https://penguindriver.com/cite?path=%2Fs%2Fai.responsibleailabs%252Frail-score&format=text",
  "how": "下載 text_snapshot 後計算 SHA-256，應與 content_sha256 相同；內容更新時指紋會改變"
 },
 "sources": [
  {
   "name": "https://github.com/Responsible-AI-Labs/rail-score-mcp",
   "url": "https://github.com/Responsible-AI-Labs/rail-score-mcp"
  },
  {
   "name": "https://responsibleailabs.ai",
   "url": "https://responsibleailabs.ai"
  }
 ],
 "source_count": 2,
 "cite_as": {
  "zh": "ai.responsibleailabs/rail-score｜台灣 AI 代理人任務中樞。https://penguindriver.com/s/ai.responsibleailabs%2Frail-score（資料日期 未標示，擷取 2026-10-17）",
  "apa": "台灣 AI 代理人任務中樞. (2026). ai.responsibleailabs/rail-score. Retrieved 2026-10-17, from https://penguindriver.com/s/ai.responsibleailabs%2Frail-score",
  "markdown": "[ai.responsibleailabs/rail-score](https://penguindriver.com/s/ai.responsibleailabs%2Frail-score)（台灣 AI 代理人任務中樞，2026-10-17）"
 },
 "license": "可引用，請附上正式網址與資料日期"
}