Customize or Remove Injected Prompt Text in Text Classifier Node

nembdev · March 10, 2026, 4:11pm

AI → Text Classifier

The idea is:

Currently the Text Classifier node injects a massive 700 token explanation of what JSON is behind the scenes. This causes unneeded token bloat, which increases api costs. 100 tokens vs 1000 tokens. The custom prompt option does not change this. Ideally we should be able to customize the entire prompt or optionally remove the injected prompt. Being optional would help avoid issues from basic use or people who don’t care.

My use case:

Saving on api costs. Node uses additional 700+ tokens due to injected prompt vs similar use case with Message A Model under 200.

I think it would be beneficial to add this because:

Saving on API costs.

Any resources to support this?

Are you willing to work on this?

??? Yall have a valuation of 2B

nembdev · March 10, 2026, 4:20pm

Another user found that this is most likely coming from langchain:

github.com/langchain-ai/langchainjs

libs/langchain-core/src/output_parsers/structured.ts

main


      
            getFormatInstructions(): string {
              return `You must format your output as a JSON value that adheres to a given "JSON Schema" instance.
          
          "JSON Schema" is a declarative language that allows you to annotate and validate JSON documents.
          
          For example, the example "JSON Schema" instance {{"properties": {{"foo": {{"description": "a list of test words", "type": "array", "items": {{"type": "string"}}}}}}, "required": ["foo"]}}
          would match an object with one required property, "foo". The "type" property specifies "foo" must be an "array", and the "description" property semantically describes it as "a list of test words". The items within "foo" must be strings.
          Thus, the object {{"foo": ["bar", "baz"]}} is a well-formatted instance of this example "JSON Schema". The object {{"properties": {{"foo": ["bar", "baz"]}}}} is not well-formatted.
          
          Your output will be parsed and type-checked according to the provided schema instance, so make sure all fields in your output match the schema exactly and there are no trailing commas!