FlowiseAI/Flowise · critical · Error

Azure Foundry API Key is missing in credentials.

Error message

Azure Foundry API Key is missing in credentials.

What it means

Thrown by AzureRerankRetriever.init when getCredentialParam('azureFoundryApiKey', ...) returns falsy. The retriever cannot authorize calls to Azure AI Foundry without the key, so it aborts during node initialization before building the compressor.

Source

Thrown at packages/components/nodes/retrievers/AzureRerankRetriever/AzureRerankRetriever.ts:127

                baseClasses: ['Document', 'json']
            },
            {
                label: 'Text',
                name: 'text',
                description: 'Concatenated string from pageContent of documents',
                baseClasses: ['string', 'json']
            }
        ]
    }

    async init(nodeData: INodeData, input: string, options: ICommonObject): Promise<any> {
        const baseRetriever = nodeData.inputs?.baseRetriever as BaseRetriever
        const model = nodeData.inputs?.model as string
        const query = nodeData.inputs?.query as string
        const credentialData = await getCredentialData(nodeData.credential ?? '', options)
        const azureApiKey = getCredentialParam('azureFoundryApiKey', credentialData, nodeData)
        if (!azureApiKey) {
            throw new Error('Azure Foundry API Key is missing in credentials.')
        }
        const azureEndpoint = getCredentialParam('azureFoundryEndpoint', credentialData, nodeData)
        if (!azureEndpoint) {
            throw new Error('Azure Foundry Endpoint is missing in credentials.')
        }
        const topK = nodeData.inputs?.topK as string
        const k = topK ? parseFloat(topK) : (baseRetriever as VectorStoreRetriever).k ?? 4
        const maxChunksPerDoc = nodeData.inputs?.maxChunksPerDoc as string
        const maxChunksPerDocValue = maxChunksPerDoc ? parseFloat(maxChunksPerDoc) : 10
        const output = nodeData.outputs?.output as string

        const azureCompressor = new AzureRerank(azureApiKey, azureEndpoint, model, k, maxChunksPerDocValue)

        const retriever = new ContextualCompressionRetriever({
            baseCompressor: azureCompressor,
            baseRetriever: baseRetriever
        })

View on GitHub (pinned to abe4a8601a)

Solutions

1. Create an Azure AI Foundry credential in Flowise and fill the API Key field.
2. Bind that credential to the Azure Rerank Retriever node.
3. Verify the key is non-empty and has not been rotated out.
4. Confirm the credential type exposes the azureFoundryApiKey parameter name.

Example fix

// before: no credential, or empty key
// after: in Flowise UI, create credential with azureFoundryApiKey=<your-key> and select it on the node

if (!azureApiKey) {
  throw new Error('Azure Foundry API key is required. Create an Azure AI Foundry credential and bind it.')
}

function hasAzureApiKey(cred: unknown): cred is { azureFoundryApiKey: string } {
  return !!cred && typeof (cred as any).azureFoundryApiKey === 'string' && (cred as any).azureFoundryApiKey.length > 0
}

// Config error, not retryable. Catch only to surface a user-friendly message.
try {
  await retriever.init(nodeData, input, options)
} catch (e) {
  if (e instanceof Error && /API Key is missing/.test(e.message)) {
    // show credential picker
  }
  throw e
}

When it happens

Trigger: No credential selected on the Azure Rerank Retriever node; selected credential lacks the azureFoundryApiKey field; credential saved with an empty key; wrong credential type bound.

Common situations: User created a generic Azure credential instead of the Foundry-specific one; key field left blank; credential was deleted or not shared with the chatflow.

Related errors

• Azure Foundry Endpoint is missing in credentials.
• Azure Rerank API call failed: ${error.message}
• OpenAI ApiKey not found
• Firecrawl API key not set. You can set it as FIRECRAWL_API_K
• Must specify one of "k" or "similarity_threshold".

AI-assisted analysis of FlowiseAI/Flowise@abe4a8601a (2026-08-12). Data as JSON: /api/errors/b3c53006cfc5b784. Report an issue: GitHub.