Function calling schema generator
Paste a function signature or a description and get the JSON tool definition for OpenAI, Anthropic, MCP or plain JSON Schema.
Your key stays in your browser.It goes straight to the provider, never to our servers, and we neither log nor keep it. Use a dedicated key with a spending cap, and delete it in the provider's dashboard when you are done.
Input
Result
The result will appear here.Every agent framework ends up at the same chore: turning a function into the JSON Schema a model reads before deciding to call it. The openai-agents SDK derives it from a Python signature and docstring; everywhere else you write it by hand, and the hand-written ones are where the bugs live — a required field that is optional in code, an enum typed as a free string, a description that never says the unit. Paste the function, in any language or as prose, choose the target, and this tool writes the definition in the exact envelope that target expects, with strict-mode rules applied for OpenAI, descriptions written for the model rather than for a human, and parameters a model should not control left out.
How it works
- Types are mapped to the tightest JSON Schema: Literal and enums become enum, dates get format, bounded numbers get minimum and maximum, nested objects stay nested.
- Each target gets its own envelope — type:function with strict:true for OpenAI, input_schema for Anthropic, inputSchema with behaviour annotations for MCP — so the output pastes straight into your code.
- Several functions in one paste produce one array with a tool per function; the output is valid JSON with no fences to strip.
Where your data goes
With your own key, what you type and the key go from your browser straight to the provider you choose; neither passes through Hysen Labs. In hosted mode, your text goes through our server to our provider (DeepSeek) and is billed in credits; we keep the token counts and cost of each run for billing, never the text itself or the answer. How the provider handles the text is governed by its own privacy policy.
This tool handles keys and credentials, so nothing about a run is saved, not even to your own history.
About your API key
We do not collect, store or leak your key: it lives only in this page's memory (unless you tick “remember in this tab”) and is gone when you close it. Still, treat any key you have pasted into a web page with care — create a dedicated key with a spending cap for use here, and delete or rotate it in your provider's dashboard when you are done.
What it costs
This tool is free, with no sign-in and no points.
Common questions
- What does OpenAI's strict mode change?
- With strict:true the model is guaranteed to emit arguments matching the schema, but the schema must list every property under required and set additionalProperties:false on every object; optional values are expressed as a type union with null. The generator applies those rules so the definition is accepted on the first try.
- Why are some of my parameters missing from the schema?
- Arguments like a database session, a logger, a request context or the current user's id are things the host passes in, not things a model should decide. The tool drops them and mentions the fact in the description when it matters, which mirrors how the SDKs treat injected dependencies.
- Does it produce Zod or Pydantic code?
- No, JSON only. Pass the JSON Schema output through the quicktype tool for TypeScript types, or use pydantic's create_model from the schema; both are mechanical once the schema is right.
The open-source behind it
This tool is a self-contained implementation. openai/openai-agents-python (MIT) does the same job as a library — if you need this behaviour inside your own program, start there rather than calling a web page.
openai/openai-agents-pythonAlso known as
- function calling schema generator
- openai tools json schema
- anthropic tool input_schema
- mcp tool definition
- json schema from python function
- llm tool definition generator