Back to guide/General Productivity

Structured JSON Output

GPT-4o is highly reliable at structured JSON output when given an exact schema. Adding confidence scores and source text spans makes the extraction auditable. The 'never fabricate' rule with null fallback prevents hallucinated data in extraction tasks.

gpt-4o-promptsinputTypeinputContentjsonSchema
Edit View
Prompt
Extract structured data from the following {{inputType}}:

<input>
{{inputContent}}
</input>

**Output as JSON matching this exact schema:**
\`\`\`json
{{jsonSchema}}
\`\`\`

**Extraction rules:**
- If a field is not found in the input, use null (never fabricate data)
- For {{ambiguousField}}: {{disambiguationRule}}
- Normalize {{fieldsToNormalize}} to {{normalizationFormat}}
- If multiple values exist for a single field, return as array

**Validation:**
- {{validationRules}}
- Include a "_confidence" field (0.0-1.0) for each extracted value
- Include a "_source" field with the exact text span used for extraction

Return ONLY the JSON object. No markdown, no explanation.

Variables to customize

{{inputType}}{{inputContent}}{{jsonSchema}}{{ambiguousField}}{{disambiguationRule}}{{fieldsToNormalize}}{{normalizationFormat}}{{validationRules}}

Why this prompt works

GPT-4o is highly reliable at structured JSON output when given an exact schema. Adding confidence scores and source text spans makes the extraction auditable. The 'never fabricate' rule with null fallback prevents hallucinated data in extraction tasks.

Save this prompt to your library

Organize, version, and access your best prompts across ChatGPT, Claude, and Cursor.

Related prompts

Task Decomposition Agent

Forcing the agent to plan before acting prevents premature execution and wasted steps. Explicit dependency mapping enables parallel execution and catches logical gaps early.

Tool Selection Agent

The ReAct pattern (Reason + Act) creates an explicit reasoning trace that improves tool selection accuracy. The error-handling rule prevents infinite retry loops.

Prompt Compressor

Explicitly requiring all functional requirements to be preserved prevents the model from over-compressing and losing critical instructions.

Memory Management Agent

Explicit memory read/write instructions create agents that improve over time. Categorization keeps memories organized, and the deduplication rule prevents context bloat.