ChatML Formatter
Convert a conversation into ChatML, Llama 3, Mistral, Gemma or Alpaca prompt templates.
Function Schema Generator
Write valid tool-calling JSON schemas without hand-editing braces.
Lowercase, verb first, underscores. This is the identifier your code dispatches on.
The model decides whether to call your tool from this sentence alone. Say what it does and when to use it.
Parameters
Parameter descriptions are prompt text, not documentation. "The city and state, for example San Francisco, CA" gets filled in reliably; "city" does not.
Quick Tool Overview
The Function Schema Generator writes valid tool-calling JSON schemas from a list of parameters, correctly nested and correctly quoted. The same definition is emitted in three shapes: the OpenAI tools format, the Anthropic tools format with its input schema, and bare JSON Schema for anything else.
- What it does
- Generates tool-calling JSON schemas from parameter definitions
- You give it
- A function name and its parameters with types and descriptions
- You get back
- The schema in OpenAI, Anthropic and plain JSON Schema formats
- Best for
- Function calling, agent tools and structured output
About the Function Schema Generator
Tool calling lives or dies on the schema. Get the JSON right and the model fills in arguments you can pass straight to your code; get it subtly wrong and you spend an afternoon reading validation errors. Add your parameters as rows — name, type, description, required, allowed values — and the schema is written for you, correctly nested and correctly quoted.
The same definition is emitted in three shapes: the OpenAI tools format, the Anthropic tools format with its input schema, and bare JSON Schema for anything else. One thing worth taking seriously is the description field on each parameter. The model reads those descriptions as instructions — they are the only thing telling it what a value means and when to supply it — so a parameter described as "the city, for example San Francisco, CA" gets filled in far more reliably than one described as "city".
Key features of the Function Schema Generator
Three output formats
OpenAI tools, Anthropic tools with input_schema, and bare JSON Schema, generated from one definition.
Parameters as rows
Name, type, description, required flag and allowed values are entered as fields rather than hand-edited braces.
Enums and constraints
Restrict a parameter to a fixed set of values so the model cannot invent one.
Descriptions treated as instructions
The model reads each parameter description as guidance, so "the city, for example San Francisco, CA" gets filled in far more reliably than "city".
Valid by construction
Nesting and quoting are handled for you, which removes an afternoon of reading validation errors.
How to use the Function Schema Generator
-
1
Name the function
Use a short verb-led name and describe what it does. The model decides whether to call it from this description.
-
2
Add the parameters
Add a row per parameter with its type, a real description, and whether it is required.
-
3
Constrain the values
For a fixed set of options, list the allowed values so the model cannot invent one.
-
4
Copy your format
Switch between the OpenAI, Anthropic and plain JSON Schema output and copy the one your SDK expects.
Frequently asked questions
Mostly the wrapper. OpenAI nests the definition under a function object with a parameters key; Anthropic puts name and description at the top level with an input_schema key. The JSON Schema inside is the same in both.
A great deal. The function description decides whether the model calls your tool at all, and each parameter description decides whether it fills the argument correctly. They are prompt text, not documentation, and vague ones are the usual cause of bad tool calls.
No. Mark as required only what your code genuinely cannot run without. Over-marking forces the model to invent values it does not have, which is worse than an absent field your code can default.
It restricts a parameter to a fixed list of allowed values. Use it for units, statuses and modes — anything with a closed set — and the model will pick from your list instead of inventing a variant your code does not handle.
Reviews of the Function Schema Generator
What people who use the Function Schema Generator think of it. Reviews come from visitors and are posted as written.
No ratings yet
Yours would be the first.
- 5 0 reviews rated 5 stars
- 4 0 reviews rated 4 stars
- 3 0 reviews rated 3 stars
- 2 0 reviews rated 2 stars
- 1 0 reviews rated 1 star
0 reviews
Newest first
No reviews yet
Be the first to rate the Function Schema Generator. It takes half a minute and no account.
Write the first reviewMore AI tools
Context Window Calculator
See whether your text fits a model context window, and how much room is left.
Few-Shot Example Builder
Turn input and output pairs into a clean, consistent few-shot prompt block.
ChatGPT Export to Markdown
Turn conversations.json from a ChatGPT or Claude export into readable Markdown files.
Fine-Tuning Dataset Formatter
Turn question and answer pairs into a JSONL training file for OpenAI, Llama or Gemini.
Embedding Cost Calculator
Estimate the cost and vector storage of embedding a document corpus for RAG.
OpenAI API Request Builder
Build a chat API request for OpenAI, Claude, Gemini, Groq or Ollama in cURL, Python, JS or PHP.
Token Counter
Estimate how many tokens your text costs before you send it.