Naive cross-leaflet drug interaction screening

tw_check_drug_interaction

TW-specificMIT

**關鍵字 / Keywords**: 藥品交互作用多藥檢查 drug interaction check polypharmacy 交互作用

初步篩查多藥品之間的交互作用 — naive substring scan over each
drug's 交互作用 leaflet field (twinkle-ai/tw-drug-labels-vision,
CC-BY-4.0).

對打 Taiwan-Health-MCP check_medical_conflict, 但只用仿單文字, 不用
SNOMED / RxNorm 知識庫. **不是臨床決策工具**, 只供初步參考.

Args:
    license_nos: 2..10 個許可證字號 list
                 e.g. ["衛署藥製字第046823號", "衛署藥輸字第023624號"]

Returns:
    {
      n_drugs, n_with_leaflet, missing_leaflet,
      potential_interactions: [{a_license, a_name, b_license, b_name,
                                matched_term, evidence_in_a}, ...],
      disclaimer
    }

如果 leaflet 沒提到對方藥名, 即使臨床上有交互作用也偵測不到 (limitation).

Input schema

{
  "properties": {
    "license_nos": {
      "items": {
        "type": "string"
      },
      "title": "License Nos",
      "type": "array"
    }
  },
  "required": [
    "license_nos"
  ],
  "title": "tw_check_drug_interactionArguments",
  "type": "object"
}

How to call

Once your Twinkle Hub MCP client is configured, your agent will see tw_check_drug_interaction. Just ask it to call — example:

# Ask Claude / any MCP client:
請用 tw_check_drug_interaction 處理 "…"。

# It will call:
tw_check_drug_interaction(input="…")

Haven't set up a client yet?

Claude Desktop in 3 minutes — download the .mcpb and double-click, or see docs for other clients.

See user docs