Help: I want to connect an external Knowledge of RagFlow. I added the required parameters by dify. How to define the required parameters by external Knowledge?

[simonjhy]

Self Checks

• I have searched for existing issues search for existing issues, including closed ones.
• I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).
• [FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:)
• Please do not modify this template :) and fill in all the required fields.

1. Is this request related to a challenge you're experiencing? Tell me about your story.

I want to connect an external Knowledge of RagFlow. I added the required parameters by dify. How to define the required parameters by external Knowledge?

below is the external Knowledge required visit api format.

curl --request POST
--url http://{address}/api/v1/retrieval
--header 'Content-Type: application/json'
--header 'Authorization: Bearer <YOUR_API_KEY>'
--data '
{
"question": "What is advantage of ragflow?",
"dataset_ids": ["b2a62730759d11ef987d0242ac120004"],
"document_ids": ["77df9ef4759a11ef8bdd0242ac120004"],
"metadata_condition": {
"logic": "and",
"conditions": [
{
"name": "author",
"comparison_operator": "=",
"value": "Toby"
},
{
"name": "url",
"comparison_operator": "not contains",
"value": "amd"
}
]
}
}'

2. Additional context or comments

No response

[dosubot[bot]]

When connecting RagFlow as an external knowledge base in Dify, you only need to configure the API endpoint and API key—Dify will forward your query to RagFlow, and RagFlow handles the retrieval logic. Dify does not natively support metadata-based filtering or advanced parameter mapping for external knowledge bases; it simply passes parameters like question, dataset_ids, document_ids, and metadata_condition to RagFlow as part of the request payload. Whether metadata_condition works depends entirely on RagFlow's API implementation, not Dify's—Dify does not enforce or interpret these filters itself. If you need advanced filtering, you can use Dify's HTTP node to call RagFlow's API directly with custom parameters, or use a gateway (like nginx+lua) to adapt request/response formats if needed. Make sure the API response from RagFlow matches Dify's expected structure: a JSON object with a records array, each containing content, score, title, and optionally metadata. For more details, see this discussion and integration notes.

To reply, just mention @dosu.

---

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}