Design, refine, and debug prompts for OpenAI GPT-5 series models.
npx skills add https://github.com/ijindal1/jaunt --skill openai-gpt5-prompting
Please help me install this Agent Skill. Skill: openai-gpt5-prompting Source: https://github.com/ijindal1/jaunt/tree/main/.codex/skills/openai-gpt5-prompting Command: npx skills add https://github.com/ijindal1/jaunt --skill openai-gpt5-prompting
This Skill helps users write, critique, and iterate on prompts for OpenAI's GPT-5 series models, ensuring clear instructions and desired model behaviors.
Use the openai-gpt5-prompting skill to draft a system message for a GPT-5 model that needs to act as a code reviewer.
High-intent search queries and answers about installing and using this skill.
To get valid JSON from GPT-5, you need to craft precise system and developer messages that define strict schema constraints. This Skill helps design structured output prompts, ensuring the model generates JSON without extra prose or format drift.
Debugging GPT-5 prompt failures involves identifying ambiguous instructions and refining message templates. This Skill helps analyze prompt failures like hallucinations or format drift, allowing you to iterate on requirements and fix unexpected model behaviors.
Controlling reasoning depth in GPT-5 prompt engineering requires tuning behavior settings within your prompts. This Skill assists in selecting between quick and deep reasoning modes by refining your system and user messages for optimal performance.
Yes, developer messages are essential for formatting tool calls in GPT-5. This Skill leverages system, developer, and user message templates to address structured output generation, ensuring your prompts correctly instruct the model on tool usage.
Yes, you can use this Skill to clarify ambiguous requirements before generating OpenAI prompts. It addresses requirement clarification as a core task, helping you develop unambiguous, testable, and robust prompts for GPT-5 models.
System messages alone may not prevent format drift or hallucinations in GPT-5 without proper evaluation design. This Skill helps identify these limitations by debugging prompt failures and designing comprehensive evaluations to test model behaviors.
The-InterdependencyRoute GPT-5 API calls across models via Responses API or Chat Completions.
Blake-JohnWrite, refactor, and evaluate LLM prompts with JSON schemas.
kimrejstromCentralize GPT-4 calls with Pydantic structured outputs in Alpacalyzer.
spideynoloveEvaluate LLM outputs with automated metrics and human evaluation checklists.
