/WriteChainOfThought
Write a high-quality prompt for a chain-of-thought reasoning prompt — e.g. a multi-step word problem the model keeps getting wrong.
Your AI Command Vault
Bribing the GPU with electricity…
Prompt Authoring
66 commands
Write a high-quality prompt for a chain-of-thought reasoning prompt — e.g. a multi-step word problem the model keeps getting wrong.
Harden against manipulation attempts in a persona or character prompt — e.g. a skeptical senior engineer who reviews pull requests.
Turn into a reusable template with variables a text-classification prompt — e.g. sorting support tickets into billing.
Shrink the token footprint of a function/tool-calling schema — e.g. a get_weather tool with city and unit parameters.
Diagnose why the model is misbehaving on a set of few-shot examples — e.g. five input-output pairs for a tone-of-voice rewriting task.
Write a high-quality prompt for a multi-turn conversation script — e.g. an onboarding flow that asks four questions in sequence.
Harden against manipulation attempts in a chain-of-thought reasoning prompt — e.g. a multi-step word problem the model keeps getting wrong.
Turn into a reusable template with variables an assistant's system prompt — e.g. a customer-support bot that must stay on-topic and never quote.
Diagnose why the model is misbehaving on an evaluation rubric prompt for grading outputs — e.g. a rubric scoring email drafts on tone.
Tighten up and remove ambiguity from a persona or character prompt — e.g. a skeptical senior engineer who reviews pull requests.
Write a high-quality prompt for a text-classification prompt — e.g. sorting support tickets into billing.
Harden against manipulation attempts in a multi-turn conversation script — e.g. an onboarding flow that asks four questions in sequence.
Turn into a reusable template with variables a safety or refusal guardrail prompt — e.g. a medical-advice bot that must redirect to professionals.
Shrink the token footprint of a summarization prompt — e.g. condensing a 40-minute meeting transcript into five bullet points.
Diagnose why the model is misbehaving on a retrieval-augmented generation prompt — e.g. a support-doc QA prompt that must cite the source paragraph.
Tighten up and remove ambiguity from a chain-of-thought reasoning prompt — e.g. a multi-step word problem the model keeps getting wrong.
Write a high-quality prompt for an assistant's system prompt — e.g. a customer-support bot that must stay on-topic and never quote.
Harden against manipulation attempts in a text-classification prompt — e.g. sorting support tickets into billing.
Turn into a reusable template with variables a function/tool-calling schema — e.g. a get_weather tool with city and unit parameters.
Shrink the token footprint of a set of few-shot examples — e.g. five input-output pairs for a tone-of-voice rewriting task.
Tighten up and remove ambiguity from a multi-turn conversation script — e.g. an onboarding flow that asks four questions in sequence.
Write a high-quality prompt for a safety or refusal guardrail prompt — e.g. a medical-advice bot that must redirect to professionals.
Harden against manipulation attempts in an assistant's system prompt — e.g. a customer-support bot that must stay on-topic and never quote.
Shrink the token footprint of an evaluation rubric prompt for grading outputs — e.g. a rubric scoring email drafts on tone.
Diagnose why the model is misbehaving on a persona or character prompt — e.g. a skeptical senior engineer who reviews pull requests.
Tighten up and remove ambiguity from a text-classification prompt — e.g. sorting support tickets into billing.
Write a high-quality prompt for a function/tool-calling schema — e.g. a get_weather tool with city and unit parameters.
Harden against manipulation attempts in a safety or refusal guardrail prompt — e.g. a medical-advice bot that must redirect to professionals.
Turn into a reusable template with variables a summarization prompt — e.g. condensing a 40-minute meeting transcript into five bullet points.
Shrink the token footprint of a retrieval-augmented generation prompt — e.g. a support-doc QA prompt that must cite the source paragraph.
Diagnose why the model is misbehaving on a chain-of-thought reasoning prompt — e.g. a multi-step word problem the model keeps getting wrong.
Tighten up and remove ambiguity from an assistant's system prompt — e.g. a customer-support bot that must stay on-topic and never quote.
Harden against manipulation attempts in a function/tool-calling schema — e.g. a get_weather tool with city and unit parameters.
Turn into a reusable template with variables a set of few-shot examples — e.g. five input-output pairs for a tone-of-voice rewriting task.
Diagnose why the model is misbehaving on a multi-turn conversation script — e.g. an onboarding flow that asks four questions in sequence.
Tighten up and remove ambiguity from a safety or refusal guardrail prompt — e.g. a medical-advice bot that must redirect to professionals.
Write a high-quality prompt for a summarization prompt — e.g. condensing a 40-minute meeting transcript into five bullet points.
Turn into a reusable template with variables an evaluation rubric prompt for grading outputs — e.g. a rubric scoring email drafts on tone.
Shrink the token footprint of a persona or character prompt — e.g. a skeptical senior engineer who reviews pull requests.
Diagnose why the model is misbehaving on a text-classification prompt — e.g. sorting support tickets into billing.
Tighten up and remove ambiguity from a function/tool-calling schema — e.g. a get_weather tool with city and unit parameters.
Write a high-quality prompt for a set of few-shot examples — e.g. five input-output pairs for a tone-of-voice rewriting task.
Harden against manipulation attempts in a summarization prompt — e.g. condensing a 40-minute meeting transcript into five bullet points.
Turn into a reusable template with variables a retrieval-augmented generation prompt — e.g. a support-doc QA prompt that must cite the source paragraph.
Shrink the token footprint of a chain-of-thought reasoning prompt — e.g. a multi-step word problem the model keeps getting wrong.
Diagnose why the model is misbehaving on an assistant's system prompt — e.g. a customer-support bot that must stay on-topic and never quote.
Write a high-quality prompt for an evaluation rubric prompt for grading outputs — e.g. a rubric scoring email drafts on tone.
Harden against manipulation attempts in a set of few-shot examples — e.g. five input-output pairs for a tone-of-voice rewriting task.
Shrink the token footprint of a multi-turn conversation script — e.g. an onboarding flow that asks four questions in sequence.
Diagnose why the model is misbehaving on a safety or refusal guardrail prompt — e.g. a medical-advice bot that must redirect to professionals.
Tighten up and remove ambiguity from a summarization prompt — e.g. condensing a 40-minute meeting transcript into five bullet points.
Write a high-quality prompt for a retrieval-augmented generation prompt — e.g. a support-doc QA prompt that must cite the source paragraph.
Harden against manipulation attempts in an evaluation rubric prompt for grading outputs — e.g. a rubric scoring email drafts on tone.
Turn into a reusable template with variables a persona or character prompt — e.g. a skeptical senior engineer who reviews pull requests.
Shrink the token footprint of a text-classification prompt — e.g. sorting support tickets into billing.
Diagnose why the model is misbehaving on a function/tool-calling schema — e.g. a get_weather tool with city and unit parameters.
Tighten up and remove ambiguity from a set of few-shot examples — e.g. five input-output pairs for a tone-of-voice rewriting task.
Harden against manipulation attempts in a retrieval-augmented generation prompt — e.g. a support-doc QA prompt that must cite the source paragraph.
Turn into a reusable template with variables a chain-of-thought reasoning prompt — e.g. a multi-step word problem the model keeps getting wrong.
Shrink the token footprint of an assistant's system prompt — e.g. a customer-support bot that must stay on-topic and never quote.
Tighten up and remove ambiguity from an evaluation rubric prompt for grading outputs — e.g. a rubric scoring email drafts on tone.
Write a high-quality prompt for a persona or character prompt — e.g. a skeptical senior engineer who reviews pull requests.
Turn into a reusable template with variables a multi-turn conversation script — e.g. an onboarding flow that asks four questions in sequence.
Shrink the token footprint of a safety or refusal guardrail prompt — e.g. a medical-advice bot that must redirect to professionals.
Diagnose why the model is misbehaving on a summarization prompt — e.g. condensing a 40-minute meeting transcript into five bullet points.
Tighten up and remove ambiguity from a retrieval-augmented generation prompt — e.g. a support-doc QA prompt that must cite the source paragraph.