A collection of advanced meta-prompts designed to generate, optimize, and evaluate other prompts. These prompts are crafted to enhance agentic behaviors, structured output, and reasoning capabilities in Large Language Models.
Prompts that guide you to the right framework for your specific needs.
| Prompt Name | Version | Description |
|---|---|---|
| Meta-Prompt Framework Selector | v3 | Analyzes your intent and selects the perfect framework from the library. |
Prompts focused on creating high-quality, agentic, and structured outputs.
| Prompt Name | Version | Description |
|---|---|---|
| Agentic Prompt Generator | v5 | Generates prompts for agentic, tool-using, and methodically-reflective LLMs. |
| Chain of Thought (CoT) Generator | v1 | Designs prompts that encourage step-by-step reasoning and logic. |
| C.R.I.S.P.E. Prompt Generator | v2 | Framework for Capacity, Role, Insight, Statement, Personality, and Experiment. |
| Few-Shot Data Synthesizer | v1 | Generates synthetic data examples for few-shot learning and context. |
Frameworks and structures for defining comprehensive and robust prompts.
| Prompt Name | Version | Description |
|---|---|---|
| C.O.S.T.A.R. Prompt Generator | v2 | Creates structured prompts using the Context, Objective, Style, Tone, Audience, Response framework. |
| C.R.A.F.T. Prompt Generator | v4 | Sophisticated prompt generation using the CRAFT methodology for high-quality interactions. |
| Deep Research Prompt Generator | v6 | Specialized for conducting deep, multi-step research tasks. |
| S.C.O.P.E. Prompt Generator | v2 | Scope, Context, Origin, Purpose, Audience framework generator. |
Tools for optimizing, evaluating, and refining existing prompts.
| Prompt Name | Version | Description |
|---|---|---|
| AI Mistake Detection & Classification | v4 | Detects and classifies errors in AI outputs to improve reliability. |
| Dynamic Variable Optimizer | v2 | Generalizes and optimizes structured prompts using dynamic variables for reusability. |
| Prompt Compressor & Token Optimizer | v1 | Compresses prompts to save tokens while maintaining core intent and instructions. |
| Prompt Evaluator | v4 | Critiques and scores prompts to identify weaknesses and areas for improvement. |
| R.T.F. Prompt Generator | v2 | Role, Task, Format framework for simple yet effective prompt structure. |
Helper prompts for format conversion and other utility tasks.
| Prompt Name | Version | Description |
|---|---|---|
| Template Converter (JSON/Jinja2) | v2 | Converts natural language prompts into structured JSON or Jinja2 templates. |
This repository also contains a full import of the prompt collection in JSON format, ready for import into PromptHive.
- File:
prompthiveJson/meta_prompts.json - Usage: You can import this file directly into your PromptHive instance to restore or duplicate the entire "Meta Prompts" collection with all its metadata, tags, and structure preserved.
Select a prompt from the collection above to help you design, refine, or analyze your own LLM interactions. Each file contains detailed instructions and context settings.
Feel free to browse, use, and adapt these prompts for your own projects.
