Welcome to the KOLO_AI™ Help Center. If you are
experiencing technical, billing or any issue that isn't answered by the below guide
please contact us at support@kolodev.com
We monitor our services for any issues but if you have a question, comment or
request feel free to reach out to us by email or directly on social media.
Coming Soon
We will shortly be allowing registered users to raise support queries directly
through their subscription interface so keep an eye out for that as well.
In this guide, we'll provide you with all the information you need to make the most of our AI-powered Assistants platform. Whether you're a beginner or an advanced user, this guide will help you navigate through KOLO_AI™'s features and capabilities.
KOLO_AI® is an advanced AI platform that leverages OpenAI's ChatGPT technology to empower users to create and customize AI Assistants for a wide range of applications. Our platform enables you to design and manage system prompts, upload and utilize documents, and access a library of pre-made prompts for various use cases.
When you first sign in you will be presented with the AI Assistants home Page. When you create an AI assistant this is where you will find them. You will be able to launch Edit, Clone and Delete your AI assistants from this page. To begin Click the "CREATE +" button.
After Clicking Create you will be given the option to Name and Select an Avatar for your AI. The Avatar is the picture used to represent your AI Assistant when you are interacting with it.
After Clicking Create you will be given the option to Name and Select an Avatar for your AI. The Avatar is the picture used to represent your AI Assistant when you are interacting with it.
Once you have named your AI and selected an icon or avatar Click "Submit" to save your AI.
After saving the AI Assistant you can now specify the system prompt for your AI. This is where you tell the AI its purpose. You can use a specific style of prompting which is covered in this help file, such as the CRISP or SMART prompt techniques. For now let's use something relatively simple to get started.
Click the Modify button to begin editing your system prompt.
This is the Prompt Editor. You can specify your prompt and
name it accordingly within this screen. Type in a prompt like the following.
"You are a helpful AI Assistant who works for us to answer queries on the data we have
uploaded to assist us in creating content and responses for our businesses and
projects."
Then click Submit to save the prompt.
Click the files link at the top to open the files tab for file uploading.
You should have a file folder dialog open and you can select one or many PDF files to upload to your AI. Files uploaded to your AI are inserted into its memory so it can use these files to answer your questions and assist you with your tasks.
Scroll to the bottom and click Save AI to save your configuration changes.
You should be returned to your AI select screen.
KOLO_AI® allows you to specify system prompts for your AI
Assistants.
There are a number of structured methods for getting the most out of your AI Assistants.
The CRISP and SMART prompt techniques are covered in this section.
We will provide Prompts in the prompt library section below for you to use within your
AI Assistants as we test and prove them out.
Feel free to take any of these prompts and use them in your AI Assistants.
Below is an informational video helping you begin your journey into AI prompt
engineering.
When engaging with AI models like KOLO_AI or others, crafting well-structured prompts is crucial to receive meaningful responses. The CRISP formula, devised by RushTree Business Optimization, provides a framework for creating effective prompts that yield better AI results. Let's break down the CRISP formula:
Your prompt should be clear about the desired AI task. Ask specific questions and provide unambiguous instructions.
Example: "Compose a 200-word synopsis of the impact of renewable energy sources on reducing greenhouse gas emissions."
Furnish relevant context and background information to contextualize the prompt. Offer the AI the necessary details to generate an appropriate response.
Example: "Considering I'm planning a trip to Tokyo and want to explore the city's history, provide an overview of the historical significance of Meiji Shrine."
Include any information or specifics the AI needs to complete the task effectively. This might involve defining terms, specifying parameters, or setting constraints.
Example: "Draft a technical report on the benefits and drawbacks of quantum computing in financial forecasting, using at least three reputable sources for references."
Ensure that your prompt addresses specific topics and provides clear, well-defined instructions. Avoid vague or open-ended queries that may result in ambiguous responses.
Example: "What are the top three strategies for improving time management for remote workers in the IT industry?"
Employ courteous and non-demanding language in your prompts. Using polite expressions such as "please" and "thank you" can enhance the quality of AI responses.
Example: "Could you kindly offer some advice on creating an effective resume for a recent college graduate seeking a marketing internship?"
By applying the CRISP approach to your prompts, you can anticipate more accurate and informative responses from AI assistants. Structuring prompts to be clear, relevant, informative, specific, and polite guides the AI in delivering valuable information. With practice, you'll become adept at formulating prompts that facilitate productive interactions with AI models.
The CRISP formula serves as a valuable tool for optimizing your interactions with AI models, whether you're seeking information, generating content, or solving specific tasks.
The SMART AI prompting technique is an approach that emphasizes specific, measurable, achievable, relevant, and time-bound (SMART) qualities in formulating prompts for AI systems. This technique helps users create effective and well-structured queries for AI models to obtain the most accurate and useful responses. Let's break down each element of the SMART AI prompting technique:
Specificity is key to SMART prompting. Your query should be clear, precise, and narrowly focused. Avoid vague or overly broad questions. This helps the AI understand exactly what you want.
Example: Instead of asking, "Tell me about technology," a specific prompt would be, "Explain the impact of artificial intelligence on the healthcare industry."
A measurable prompt should have a clearly defined outcome or result. It allows you to assess the success of the AI's response based on specific criteria.
Example: "List the top five features of the latest smartphone model" is measurable because you can easily compare the AI's response to the expected five features.
An achievable prompt ensures that the AI can reasonably fulfill your request. It should be within the AI's capabilities and resources.
Example: "Translate this English text into French" is achievable if the AI has language translation capabilities. However, "Predict the stock market accurately" might not be achievable due to the complexity of the task.
Relevance in AI prompts means that your query should be related to the AI model's expertise or the context of your interaction. Irrelevant questions may result in inaccurate or off-topic responses.
Example: If you're interacting with a weather forecasting AI, asking, "What's the best restaurant in town?" is not relevant to its capabilities.
A time-bound prompt specifies a timeframe for the AI to consider. It can be important for questions that involve historical data, future predictions, or time-sensitive information.
Example: "Provide a summary of technological advancements in the last decade" sets a time boundary and context for the AI's response.
In summary, the SMART AI prompting technique encourages users to create prompts that are specific, measurable, achievable, relevant, and time-bound. This approach enhances the clarity and effectiveness of interactions with AI systems, resulting in more accurate and valuable responses tailored to your needs and objectives.
The Zero-Shot Prompting technique allows for generating responses without providing any prior examples or context. It is ideal for obtaining quick answers to basic questions or general topics.
Prompt: "What is the capital of France?"
AI Response: "The capital of France is Paris."
- Quick and straightforward, no need for prior examples or context.
- Suitable for basic questions or general topics.
- Minimal user input required.
- Limited to providing factual or basic information.
- May not handle complex or nuanced queries well.
- General knowledge questions, e.g., trivia games.
- Quick information retrieval for common facts.
- Basic customer support inquiries in various industries.
One-Shot Prompting is an approach that extracts responses based on a single example or piece of context provided by the user. It simplifies interactions with AI by requiring just one relevant prompt for generating responses.
Example: User provides the prompt, "Explain the concept of renewable energy sources."
- Simplifies the interaction by requiring only one example or context.
- Useful for obtaining specific information or explanations quickly.
- Limited to generating responses based on a single input.
- May not handle complex or multifaceted queries effectively.
- Answering questions based on a specific context provided by the user.
- Providing explanations or information related to a single prompt.
- Assisting users in getting quick responses to straightforward queries.
Few-shot prompting refers to presenting a language model with a task or question along with a few examples of the desired output. This technique helps guide the model in the right direction.
Example: Identify suitable features for a project management software
based on the given problems:
Problem 1:
Teams have difficulty in tracking the progress of their projects.
Feature 1:
Progress tracking dashboard with visual indicators and real-time updates.
- Helps guide the model with examples.
- Useful for tasks requiring reasoning.
- Still requires manual input.
- Limited to the provided examples.
- Idea generation.
- Task guidance.
Crafting effective prompts is a skill essential for obtaining precise and valuable responses from AI models like ChatGPT, Google Bard, Claude, and others. The success of your interactions with these models hinges on how well you construct your prompts. Let's delve into six key components of prompt construction, each with its own unique importance, examples, and advantages and disadvantages.
These six components play pivotal roles in crafting impactful prompts and are listed in order of descending significance. Let's explore each component in detail with original examples and the benefits and limitations:
Importance: The task provides the primary goal for AI responses.
Example: "Generate a compelling marketing tagline for our new product."
Pros: Precise instructions. Promotes clear responses.
Cons: May yield concise answers that lack context.
Importance: Context frames the AI's understanding and relevance of the task.
Example: "Our company is launching a sustainable fashion line, and we need a catchy tagline that reflects our eco-friendly mission."
Pros: Ensures AI is well-informed. Enhances response quality.
Cons: Can make prompts more complex and lengthy.
Importance: Including examples in prompts helps AI understand and improve responses.
Example: "Here's a sample tagline for reference: 'Style with a Green Heart.' Can you provide similar options?"
Pros: AI learns from concrete examples. Results are more focused.
Cons: Not needed for all prompts. May lengthen prompts unnecessarily.
Importance: Specifies the role or style of communication for AI responses.
Example: "Respond in a friendly, casual tone, similar to how a helpful neighbor would."
Pros: Tailors responses to specific situations. Enhances relatability.
Cons: Limited to well-known personas. Might not always be relevant.
Importance: Structures information for clarity and efficiency in AI responses.
Example: "Please provide the answer in a bulleted list format."
Pros: Organizes responses for better understanding. Ideal for specific use cases.
Cons: May not suit all prompts. Could restrict creative responses.
Importance: Specifies the style and tone of language in AI responses.
Example: "Respond in a formal and professional tone, as if addressing a corporate audience."
Pros: Tailors responses to match the desired communication style. Ideal for specific audiences.
Cons: May not always be relevant. Might stifle creative expression.
Skillfully combining these components in your prompts guarantees that AI generates high-quality, customized responses. The choice of which elements to include should be guided by the nature of your query. By grasping the importance of each component, you can construct prompts that make the most of AI systems.
Information Retrieval prompting treats large language models as search engines. It involves asking highly specific questions to obtain detailed answers. The effectiveness depends on the model's data sources.
Example: User asks, "What are the latest trends in artificial intelligence?"
- Ideal for obtaining specific and detailed information.
- Effective for staying updated on recent trends or data.
- Limited by the AI's data sources and knowledge cutoff date.
- May not handle general or abstract queries as effectively.
- Staying informed about recent developments in specific fields.
- Gathering detailed data or information for research purposes.
- Accessing specific facts or statistics from reliable sources.
Creative Writing involves crafting prompts to generate imaginative narratives, captivating stories, and unique textual content tailored to your audience's preferences and interests.
Example: User prompts the AI, "Write a short story about a detective solving a mysterious case in a small town."
- Enables the generation of engaging and creative content.
- Tailored content for various storytelling and audience needs.
- May require clear and detailed prompts for the desired creative output.
- Quality and relevance may vary based on the AI's capabilities.
- Content creation for fiction stories, marketing, or entertainment.
- Generating engaging narratives for websites, blogs, or books.
- Customizing content for specific storytelling needs.
Context Expansion involves enriching the information provided to the AI to enhance its understanding effectively. It is achieved by asking "Who, What, Where, When, Why, and How" questions related to the subject matter.
Example: Expanding the context of a simple statement, "Exercise is good for your health," by asking additional questions about why, how, and where to exercise.
- Enhances the AI's understanding by providing comprehensive context.
- Helps extract detailed and well-rounded information from the AI.
- Requires the formulation of multiple context-expanding questions.
- May increase complexity in the prompt but results in richer responses.
- Gathering comprehensive information on specific topics or subjects.
- Creating detailed content for research or educational purposes.
- Expanding understanding on complex subjects through detailed inquiries.
Content Summarization with Specific Focus is about directing AI's attention to specific parts of the input, especially when you want concise summaries with a particular emphasis. It involves crafting precise instructions to highlight which elements of the text to prioritize.
Example: "Summarize this article on website optimization, but only focus on strategies related to mobile optimization."
- Allows for the creation of tailored, concise summaries.
- Ensures that the summary captures the essence of the desired focus.
- Requires clear and specific instructions to achieve the desired focus.
- Quality of the summary may depend on the AI's capabilities.
- Creating targeted summaries for research papers or articles.
- Generating focused content for specific web pages or reports.
- Customizing content for a particular audience or purpose.
Template Filling allows you to create versatile yet structured content effortlessly. You use a template with placeholders to enable prompt customization for different situations or inputs while maintaining a consistent format.
Example: "Generate a personalized welcome email using the template 'Hello {Name}, Welcome to our {Service}. We're glad you're here! {Closing}'"
- Allows for the creation of structured content that can be customized for various
situations.
- Ensures consistent format while adapting to specific inputs.
- Requires template setup and clear variable definitions.
- The quality of the generated content may depend on the AI's capabilities.
- Creating consistent product descriptions for e-commerce websites.
- Generating automated but personalized email content for marketing campaigns.
- Structuring content for various templates and formats.
Prompt Reframing or AI prompt customization lets you subtly change your prompts' wording while maintaining the query's original intent. It can encourage the language model to produce a variety of responses that answer the original intent in different ways.
Example: Original Prompt – "What are some ways to optimize a website
for speed?"
Reframed Prompt – "Can you list techniques to improve website loading times?"
- Allows for the generation of a variety of responses based on the same intent.
- Useful for exploring different ideas or perspectives on a topic.
- Requires skill in prompt phrasing to achieve the desired variations.
- The quality of responses may vary based on the AI's capabilities.
- Exploring different angles of a topic or issue for creative content.
- Generating multiple solutions or ideas for a specific problem.
- Crafting diverse responses for a natural and engaging conversation.
The Prompt Combination technique involves merging different instructions or questions into a single, multi-faceted prompt to elicit a comprehensive answer from the AI.
Example: User asks, "Can you explain the differences between shared hosting and VPS hosting and recommend which is better for a small eCommerce website?"
- Enables the extraction of comprehensive responses to multi-part questions.
- Useful for obtaining well-rounded answers that cover various aspects of a topic.
- Requires the formulation of complex, multi-faceted prompts.
- Response quality may vary depending on the AI's capabilities.
- Obtaining detailed explanations or comparisons for complex topics.
- Seeking comprehensive recommendations or answers to multifaceted questions.
- Navigating in-depth discussions on various subjects.
Chain-Of-Thought Prompting is about using real-time AI interactions to guide the tool toward more accurate and comprehensive answers. It involves providing a sequence of related examples or questions to refine the original query.
Example: Breaking down a complex question into smaller, more manageable parts and presenting these parts as a sequence of prompts to refine the query.
- Allows for in-depth exploration of complex topics or questions.
- Refines the AI's responses through a sequence of related prompts.
- Requires careful planning and sequencing of prompts for best results.
- The quality of responses may vary based on the AI's capabilities.
- Gathering detailed information or insights on multifaceted subjects.
- Clarifying ambiguities, extracting additional insights, or exploring complex topics in
depth.
- Guiding AI to provide comprehensive answers to complex questions.
Iterative Prompting is an effective technique where you build upon previous responses by asking follow-up questions. This allows you to dive deeper into a topic, extract additional insights, or clarify any ambiguities from the initial output.
Example: Framing a follow-up prompt to elaborate on a specific part of the initial response, dig into a subtopic, or seek clarification.
- Enables in-depth exploration and clarification of topics or responses.
- Extracts additional insights and details through follow-up questions.
- Relies on the AI's ability to handle follow-up questions effectively.
- May require multiple interactions to gather detailed information.
- Gathering detailed information on complex or multifaceted topics.
- Clarifying ambiguities, exploring subtopics, or seeking additional insights.
- Conducting interviews or interactions that require follow-up questions.
Interactive Storytelling and Role-Playing with AI is a creative and engaging way to produce dynamic content. This technique takes advantage of the AI's ability to adapt its responses based on the prompts and previous interactions, allowing for a fluid and evolving narrative.
Example: Prompt – "I want to start a collaborative storytelling exercise with you. We'll be writing a fantasy story about a land where magic exists but has been forgotten for a long time. Here's the main character and the end goal of the story."
- Engages users in creative and dynamic content creation.
- Allows for collaborative storytelling and evolving narratives.
- Quality of the storytelling may depend on the AI's capabilities.
- May require ongoing input and engagement to maintain the narrative.
- Collaborative storytelling exercises for creative writing or entertainment.
- Interactive and evolving narratives for gaming or role-playing scenarios.
- Creating engaging and dynamic content for diverse audiences.
With Implicit Information Injection, you infuse context subtly so the AI model understands your needs without being explicitly told. This technique can be particularly useful when you're seeking nuanced or tailored responses.
Example: Prompt – "Can you give some quick tips for boosting website traffic?"
- Enables the AI to provide nuanced or tailored responses.
- Useful for achieving specific tones, styles, or contextual understanding.
- Requires skill in prompt phrasing to achieve the desired nuances.
- Response quality may vary depending on the AI's capabilities.
- Tailoring responses for different writing styles, tones, or cultural nuances.
- Crafting content that requires specific contextual understanding or implications.
- Achieving more personalized or audience-specific content.
Multilingual content generation is not just about converting words from one language to another. Using prompt engineering, you can ensure that the AI considers the cultural or situational context, resulting in a more accurate and nuanced translation.
Example: Prompt – "Translate the phrase 'She took the ball and ran with it' from English to French, keeping in mind that this is a business metaphor for taking charge of a project."
- Ensures translations consider cultural or situational context for accuracy.
- Useful for business communications, legal documents, or any text where nuance is
important.
- Requires clear contextual instructions to achieve nuanced translations.
- Response quality may vary depending on the AI's language capabilities.
- Translating business documents, legal agreements, or marketing materials with cultural
considerations.
- Ensuring translations are accurate and contextually appropriate for diverse
audiences.
- Multilingual content generation with a focus on nuanced and precise translations.
Self-consistency is a powerful prompting technique used to enhance complex reasoning tasks performed by language models. It involves a unique decoding strategy that sets it apart from conventional methods. In self-consistency, the language model doesn't rely on a single, straightforward answer. Instead, it embraces the idea that complex reasoning problems often have multiple correct solutions, each with its own unique reasoning path.
The key to self-consistency is its ability to explore a variety of reasoning paths, avoiding the limitations of traditional, one-size-fits-all approaches. This technique encourages the model to sample diverse reasoning paths and produce a set of possible answers. By doing so, it captures the richness and complexity of the problem at hand, offering a more comprehensive understanding of the subject matter.
- Enhanced reasoning for complex tasks: Self-consistency excels in scenarios where conventional methods may fall short. Its ability to explore multiple reasoning paths makes it ideal for problems that require a deep understanding and nuanced approach. - Allows for multiple reasoning paths: Instead of being confined to a single solution, self-consistency embraces diversity in reasoning. This can lead to more creative and insightful answers. - Improves the quality of answers: By sampling from various reasoning paths, self-consistency increases the likelihood of providing high-quality, well-reasoned responses that align with the problem's complexity.
- Requires careful implementation for best results: Achieving the full potential of self-consistency may require thoughtful setup and tuning to ensure the model effectively explores reasoning paths and delivers meaningful responses. - Performance depends on the model's capabilities: The effectiveness of self-consistency depends on the underlying language model. Not all models may exhibit the same level of proficiency in executing this technique.
General Knowledge Prompting is a valuable prompt engineering technique that involves augmenting a query with additional knowledge before sending the final API call to language models. This technique enriches the context of the query, enabling the language model to provide more accurate and context-aware responses.
- Enhances context: By providing additional knowledge, general knowledge prompting improves the context of the query, resulting in more accurate and insightful answers. - Boosts common sense reasoning: This technique is particularly effective in enhancing a model's ability to reason based on general knowledge, making it suitable for a wide range of tasks. - Improves the model's performance: General knowledge prompting can lead to improved overall performance, especially in tasks that require a deep understanding of the subject matter.
- Requires access to relevant knowledge sources: Effective general knowledge prompting relies on access to accurate and up-to-date knowledge sources to provide meaningful context. - Task-specific knowledge may still be necessary: While this technique enhances general knowledge reasoning, certain tasks may require task-specific knowledge that general knowledge alone cannot cover.
- Common sense reasoning tasks.
- Tasks that benefit from enriched context.
- Improving the performance of language models.
ReAct, an acronym for "Reasoning and Acting," represents an innovative approach to prompt engineering that transforms language models into versatile problem-solvers. This technique empowers language models not only to reason about complex problems but also to take relevant actions in response, creating a seamless synergy between reasoning and action.
In practical terms, ReAct allows language models to generate both reasoning traces and task-specific actions in an interleaved manner. This means that the model can not only analyze and understand complex scenarios but can also actively interact with external knowledge bases, tools, or environments to gather additional information or implement solutions.
- Comprehensive problem-solving: ReAct takes problem-solving to the next level by combining reasoning and action. This comprehensive approach is invaluable for addressing real-world challenges that require not only understanding but also practical solutions. - Interfacing with external sources: ReAct enables language models to interface with external knowledge sources and tools, making them highly adaptable to various domains and applications.
- Requires careful setup: Implementing ReAct effectively may involve careful configuration to ensure that the reasoning and action generation align with the intended task. This can be a complex process. - Dependent on model capabilities: The success of ReAct depends on the capabilities of the underlying language model. Not all models may excel in this technique.
Introduction:
To ensure that your AI system can efficiently extract text from documents, follow
these guidelines for document preparation. These practices will help your AI produce
more accurate and relevant responses.
Use descriptive and meaningful file names. A well-labeled document helps the AI locate and identify relevant information.
Include only the information that is pertinent to your queries. Unnecessary data may lead to irrelevant responses.
Avoid repetition by eliminating redundant content. Repeated information can confuse the AI and lead to duplication in responses.
Use clear and standard language to ensure the AI's understanding and accuracy.
Organize your documents logically with headings and sections for easy navigation and comprehension.
Regularly update your documents to provide the AI with the latest information and data.
Conclusion:
Following these document preparation best practices will help your AI system
efficiently extract text. This, in turn, will result in more precise and relevant
responses, enhancing the overall effectiveness of your AI system.
Ensure that your document names are descriptive and reflective of their content.
Include only relevant information and avoid repetition to assist the AI in providing accurate responses.
Use clear and standard language in your documents for better AI comprehension.
Organize your documents with headings and logical structure to enhance AI understanding.
Keep your documents up to date to ensure that the AI has access to the most current information.
In this section, we'll provide you with example comprehensive prompts.
Here, you can see examples of:
Feel free to ask questions or request information on any topic related to our application, and we'll do our best to assist you. Let's get started!
You are a versatile AI bid writing assistant, capable of crafting effective responses for bids on behalf of any company. Your primary goal is to help with bid responses by providing well-structured and informative content. Please adhere to the following guidelines: