{
  "id": 11772653,
  "title": "Your Policies Are Out of Date: How I Built a Sanity AI Agent to Catch Fact Drift",
  "url": "https://urgent.news/2026/10/03/your-policies-are-out-of-date-how-i-built-a-sanity-ai-agent-to-catch",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-03T21:21:17.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/pritam_patra_429a25dedae6/your-policies-are-out-of-date-how-i-built-a-sanity-ai-agent-to-catch-fact-drift-5bee"
  },
  "original_language": "en",
  "account": "A SaaS company faced the issue of \"fact drift,\" where critical policy information like refund windows or SLA commitments became outdated across multiple pages and documents. A product manager extended the refund window from 30 days to 60 days but overlooked the fact that other pages still referenced the old 30-day value. Within weeks, a customer disputed a charge after finding the outdated pricing information, highlighting the gravity of the problem. This scenario represents a structural issue rather than a CMS problem, as the same information was stored in multiple places without any relationships between them. Fact Ledger, built on Sanity Content Lake, addresses this by storing business values as first-class Sanity documents called facts. Pages reference these facts using a custom factRef inline Portable Text annotation, allowing for automatic updates when facts change. The system implements five scanning rules to detect various types of fact drift, such as unlinked matches, contradictions, deprecated references, orphan facts, and temporal violations. The deterministic scanner runs in milliseconds, providing precise findings with exact block keys and character offsets. An AI agent then drafts the necessary Sanity patch mutations to fix the issues, which a human reviewer approves. The entire process is atomic, ensuring all updates occur in a single transaction. The system also includes a visual Clause Impact Tree to show how a single policy change ripples through the entire document graph. Overall, Fact Ledger efficiently solves fact drift by integrating fact management into Sanity's data model, preventing manual updates and ensuring consistency across all content pages.",
  "summary": "This is a submission for the Sanity Challenge, Path Two: Vibe-Code Something Strange What I Built Fact Ledger — an AI-assisted fact drift detection and remediation engine built on Sanity Content Lake. Here's the problem it solves, told as a scene: A product manager at a SaaS company decides to extend refunds from 30 days to 60 days. She opens the CMS, finds the Refund Policy page, updates it,…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}