vkc-regulation-knowledge-updater

构建监管/知识更新流水线(官方来源 → 快照 → 结构化规则集/模板 → 管理员批准 → 激活)。用于保持签证规则和文档要求最新,无需硬编码。

📖 中文导读:构建监管/知识更新流水线(官方来源 → 快照 → 结构化规则集/模板 → 管理员批准 → 激活)。用于保持签证规则和文档要求最新,无需硬编码。 以下正文为该技能的原始 SKILL.md 内容(英文),上方安装命令可直接使用。

name: vkc-regulation-knowledge-updater description: Build the regulation/knowledge update pipeline (official sources -> snapshots -> structured rulesets/templates -> admin approval -> active). Use for keeping visa rules and document requirements up-to-date without code hardcoding. metadata: short-description: Regulation knowledge updater (admin-approved)

VKC Regulation / Knowledge Updater (P2)

Goal

Keep “규정/서류 요구사항/공식 공지” up-to-date as data, not code:

  • fetch or ingest updates into snapshots
  • detect changes
  • require admin approval to activate
  • feed active rulesets into visa assessment + doc templates

Core model (minimum)

  • immigration_sources (allowlist)
  • immigration_source_snapshots (hash + fetchedAt + raw text / attachment path)
  • immigration_rulesets (structured JSON + version + status)

Admin workflow (required)

  • sync job writes pending snapshots/rulesets
  • admin reviews diffs and activates the new version

Implementation notes (practical)

  • Treat “no official API” sources as best-effort:
    • low frequency + caching + allowlist
    • immediate fallback to manual admin upload if unstable
  • Scheduler: external cron hits /api/admin/immigration/sync (authenticated)

STEP3 official sources (SoT)

  • Source list (A/B/C) and update cadence: docs/STEP3_SOT_RESOURCES.md