FlowiseAI/Flowise · error · Error

Invalid JSON in the Worker's Prompt Input Values: ${exceptio

Error message

Invalid JSON in the Worker's Prompt Input Values: ${exception}

What it means

Thrown inside Worker.init's try/catch around `JSON.parse(promptValuesStr)`. The promptValues field accepts either an object or a JSON string of substitution values for the worker prompt template; if the string is malformed JSON, parse throws and is rethrown with this message.

Source

Thrown at packages/components/nodes/multiagents/Worker/Worker.ts:106

        tools = flatten(tools)
        let workerPrompt = nodeData.inputs?.workerPrompt as string
        const workerLabel = nodeData.inputs?.workerName as string
        const supervisor = nodeData.inputs?.supervisor as IMultiAgentNode
        const maxIterations = nodeData.inputs?.maxIterations as string
        const model = nodeData.inputs?.model as BaseChatModel
        const promptValuesStr = nodeData.inputs?.promptValues

        if (!workerLabel) throw new Error('Worker name is required!')
        const workerName = workerLabel.toLowerCase().replace(/\s/g, '_').trim()

        if (!workerPrompt) throw new Error('Worker prompt is required!')

        let workerInputVariablesValues: ICommonObject = {}
        if (promptValuesStr) {
            try {
                workerInputVariablesValues = typeof promptValuesStr === 'object' ? promptValuesStr : JSON.parse(promptValuesStr)
            } catch (exception) {
                throw new Error("Invalid JSON in the Worker's Prompt Input Values: " + exception)
            }
        }
        workerInputVariablesValues = handleEscapeCharacters(workerInputVariablesValues, true)

        const llm = model || (supervisor.llm as BaseChatModel)
        const multiModalMessageContent = supervisor?.multiModalMessageContent || []

        const abortControllerSignal = options.signal as AbortController
        const workerInputVariables = getInputVariables(workerPrompt)

        if (!workerInputVariables.every((element) => Object.keys(workerInputVariablesValues).includes(element))) {
            throw new Error('Worker input variables values are not provided!')
        }

        const agent = await createAgent(
            llm,
            [...tools],
            workerPrompt,

View on GitHub (pinned to abe4a8601a)

Solutions

Provide promptValues as valid JSON: `{"key": "value"}` with double quotes.
If passing an object from a previous node, leave it as an object — the code path `typeof promptValuesStr === 'object'` skips JSON.parse.
Run the JSON through a linter/validator before pasting.
Improve the error message to include the parse position from `exception.message`.

Example fix

// before
workerInputVariablesValues = typeof promptValuesStr === 'object' ? promptValuesStr : JSON.parse(promptValuesStr)
// after
if (typeof promptValuesStr === 'string') {
  try { workerInputVariablesValues = JSON.parse(promptValuesStr) }
  catch (e) { throw new Error(`promptValues is not valid JSON at offset ${(e as Error).message}: got ${promptValuesStr.slice(0,80)}`) }
} else {
  workerInputVariablesValues = promptValuesStr
}

Defensive patterns

Strategy: validation

Validate before calling

function parsePromptValues(input: unknown): ICommonObject {
  if (input == null) return {}
  if (typeof input === 'object') return input as ICommonObject
  try { return JSON.parse(input as string) }
  catch (e) { throw new Error(`promptValues invalid JSON: ${(e as Error).message}`) }
}

Type guard

const isPlainObject = (v: unknown): v is Record<string, unknown> =>
  typeof v === 'object' && v !== null && !Array.isArray(v)

Try / catch

try { workerInputVariablesValues = JSON.parse(promptValuesStr) }
catch (e) { throw new Error(`promptValues JSON parse failed: ${(e as Error).message}`) }

Prevention

Always use double-quoted JSON.
Pass objects between nodes instead of strings when possible.
Validate JSON in the UI before submission.

When it happens

Trigger: User types a non-JSON string (e.g. `name=John`), uses single quotes, leaves a trailing comma, or pastes a JS object literal into the promptValues field.

Common situations: Manually editing promptValues in the node UI; passing a templated string that didn't get JSON-encoded upstream; locale-specific quote characters copied from a word processor.

Understand the failure class

Parsing and encoding errors: unexpected token, malformed input — why parsers reject input and how to find the real culprit.

Related errors

AI-assisted analysis of FlowiseAI/Flowise@abe4a8601a (2026-08-12). Data as JSON: /api/errors/3b038ef82aaa6506. Report an issue: GitHub.