Why do LLMs return invalid JSON?

Models sometimes wrap JSON in markdown fences, add explanatory text, or emit trailing commas. The enforced prompt explicitly forbids those, and the snippet strips fences and slices to the outermost braces before parsing as a safety net.

How does the retry loop work?

If the response fails to parse, the snippet sends a follow-up prompt that includes the exact parse error and the bad output, asking the model to return corrected JSON. It repeats up to your max-retries limit before throwing.

Does this replace function calling or structured outputs APIs?

Provider-native structured output is stronger when available. This tool is provider-agnostic and works with any model, including older ones and self-hosted models that lack a native JSON mode.

Is anything sent to a server?

No. The prompt and code snippet are generated entirely in your browser. Nothing you type is uploaded or stored.

Structured Output Enforcer

Structured output enforcer

Getting an LLM to reliably return parseable JSON is a common production headache: stray markdown fences, chatty preambles, and trailing commas all break JSON.parse. The robust pattern combines a strict prompt that bans those habits with a retry loop that feeds the parse error back to the model and asks for a correction. This tool generates both.

How it works

You provide a base prompt and the JSON schema you expect. The tool produces an enforced prompt that tells the model to return only valid JSON matching your schema — no fences, no commentary. It also emits a provider-agnostic JavaScript snippet with an extractJson helper that strips code fences and isolates the outermost object, plus a loop that, on any parse failure, re-prompts the model with the error and the bad output up to your chosen retry limit.

Tips and notes

Keep the schema small and flat. Deeply nested schemas are harder for models to honor; flatten where you can and validate nested parts separately.
Prefer native JSON mode when you have it. If your provider offers structured outputs or function calling, use it as the primary path and keep this retry loop as a fallback for other models.
Log retries. A high retry rate signals a prompt or schema problem worth fixing rather than papering over with more attempts.
Validate after parsing. Successful JSON.parse only means it is valid JSON, not that it matches your schema — add a schema check (Zod, Ajv) before trusting the data.