@promptbook/core 0.89.0 → 0.92.0-10

package/README.md

@@ -23,6 +23,10 @@
 
 
 
+<blockquote style="color: #ff8811">
+    <b>⚠ Warning:</b> This is a pre-release version of the library. It is not yet ready for production use. Please look at <a href="https://www.npmjs.com/package/@promptbook/core?activeTab=versions">latest stable release</a>.
+</blockquote>
+
 ## 📦 Package `@promptbook/core`
 
 - Promptbooks are [divided into several](#-packages) packages, all are published from [single monorepo](https://github.com/webgptorg/promptbook).

package/esm/index.es.js

@@ -27,7 +27,7 @@ const BOOK_LANGUAGE_VERSION = '1.0.0';
  * @generated
  * @see https://github.com/webgptorg/promptbook
  */
-const PROMPTBOOK_ENGINE_VERSION = '0.89.0';
+const PROMPTBOOK_ENGINE_VERSION = '0.92.0-10';
 /**
  * TODO: string_promptbook_version should be constrained to the all versions of Promptbook engine
  * Note: [💞] Ignore a discrepancy between file name and entity name
@@ -1554,7 +1554,7 @@ function extractParameterNames(template) {
  */
 function unpreparePipeline(pipeline) {
     let { personas, knowledgeSources, tasks } = pipeline;
-    personas = personas.map((persona) => ({ ...persona, modelRequirements: undefined, preparationIds: undefined }));
+    personas = personas.map((persona) => ({ ...persona, modelsRequirements: undefined, preparationIds: undefined }));
     knowledgeSources = knowledgeSources.map((knowledgeSource) => ({ ...knowledgeSource, preparationIds: undefined }));
     tasks = tasks.map((task) => {
         let { dependentParameterNames } = task;
@@ -1821,7 +1821,7 @@ function createSubcollection(collection, predicate) {
     };
 }
 
-var PipelineCollection = [{title:"Prepare Knowledge from Markdown",pipelineUrl:"https://promptbook.studio/promptbook/prepare-knowledge-from-markdown.book",formfactorName:"GENERIC",parameters:[{name:"knowledgeContent",description:"Markdown document content",isInput:true,isOutput:false},{name:"knowledgePieces",description:"The knowledge JSON object",isInput:false,isOutput:true}],tasks:[{taskType:"PROMPT_TASK",name:"knowledge",title:"Knowledge",content:"You are experienced data researcher, extract the important knowledge from the document.\n\n# Rules\n\n-   Make pieces of information concise, clear, and easy to understand\n-   One piece of information should be approximately 1 paragraph\n-   Divide the paragraphs by markdown horizontal lines ---\n-   Omit irrelevant information\n-   Group redundant information\n-   Write just extracted information, nothing else\n\n# The document\n\nTake information from this document:\n\n> {knowledgeContent}",resultingParameterName:"knowledgePieces",dependentParameterNames:["knowledgeContent"]}],personas:[],preparations:[],knowledgeSources:[],knowledgePieces:[],sources:[{type:"BOOK",path:null,content:"# Prepare Knowledge from Markdown\n\n-   PIPELINE URL `https://promptbook.studio/promptbook/prepare-knowledge-from-markdown.book`\n-   INPUT PARAMETER `{knowledgeContent}` Markdown document content\n-   OUTPUT PARAMETER `{knowledgePieces}` The knowledge JSON object\n\n## Knowledge\n\n<!-- TODO: [🍆] -FORMAT JSON -->\n\n```markdown\nYou are experienced data researcher, extract the important knowledge from the document.\n\n# Rules\n\n-   Make pieces of information concise, clear, and easy to understand\n-   One piece of information should be approximately 1 paragraph\n-   Divide the paragraphs by markdown horizontal lines ---\n-   Omit irrelevant information\n-   Group redundant information\n-   Write just extracted information, nothing else\n\n# The document\n\nTake information from this document:\n\n> {knowledgeContent}\n```\n\n`-> {knowledgePieces}`\n"}],sourceFile:"./books/prepare-knowledge-from-markdown.book"},{title:"Prepare Keywords",pipelineUrl:"https://promptbook.studio/promptbook/prepare-knowledge-keywords.book",formfactorName:"GENERIC",parameters:[{name:"knowledgePieceContent",description:"The content",isInput:true,isOutput:false},{name:"keywords",description:"Keywords separated by comma",isInput:false,isOutput:true}],tasks:[{taskType:"PROMPT_TASK",name:"knowledge",title:"Knowledge",content:"You are experienced data researcher, detect the important keywords in the document.\n\n# Rules\n\n-   Write just keywords separated by comma\n\n# The document\n\nTake information from this document:\n\n> {knowledgePieceContent}",resultingParameterName:"keywords",dependentParameterNames:["knowledgePieceContent"]}],personas:[],preparations:[],knowledgeSources:[],knowledgePieces:[],sources:[{type:"BOOK",path:null,content:"# Prepare Keywords\n\n-   PIPELINE URL `https://promptbook.studio/promptbook/prepare-knowledge-keywords.book`\n-   INPUT PARAMETER `{knowledgePieceContent}` The content\n-   OUTPUT PARAMETER `{keywords}` Keywords separated by comma\n\n## Knowledge\n\n<!-- TODO: [🍆] -FORMAT JSON -->\n\n```markdown\nYou are experienced data researcher, detect the important keywords in the document.\n\n# Rules\n\n-   Write just keywords separated by comma\n\n# The document\n\nTake information from this document:\n\n> {knowledgePieceContent}\n```\n\n`-> {keywords}`\n"}],sourceFile:"./books/prepare-knowledge-keywords.book"},{title:"Prepare Knowledge-piece Title",pipelineUrl:"https://promptbook.studio/promptbook/prepare-knowledge-title.book",formfactorName:"GENERIC",parameters:[{name:"knowledgePieceContent",description:"The content",isInput:true,isOutput:false},{name:"title",description:"The title of the document",isInput:false,isOutput:true}],tasks:[{taskType:"PROMPT_TASK",name:"knowledge",title:"Knowledge",content:"You are experienced content creator, write best title for the document.\n\n# Rules\n\n-   Write just title, nothing else\n-   Write maximum 5 words for the title\n\n# The document\n\n> {knowledgePieceContent}",resultingParameterName:"title",expectations:{words:{min:1,max:8}},dependentParameterNames:["knowledgePieceContent"]}],personas:[],preparations:[],knowledgeSources:[],knowledgePieces:[],sources:[{type:"BOOK",path:null,content:"# Prepare Knowledge-piece Title\n\n-   PIPELINE URL `https://promptbook.studio/promptbook/prepare-knowledge-title.book`\n-   INPUT PARAMETER `{knowledgePieceContent}` The content\n-   OUTPUT PARAMETER `{title}` The title of the document\n\n## Knowledge\n\n-   EXPECT MIN 1 WORD\n-   EXPECT MAX 8 WORDS\n\n```markdown\nYou are experienced content creator, write best title for the document.\n\n# Rules\n\n-   Write just title, nothing else\n-   Write maximum 5 words for the title\n\n# The document\n\n> {knowledgePieceContent}\n```\n\n`-> {title}`\n"}],sourceFile:"./books/prepare-knowledge-title.book"},{title:"Prepare Persona",pipelineUrl:"https://promptbook.studio/promptbook/prepare-persona.book",formfactorName:"GENERIC",parameters:[{name:"availableModelNames",description:"List of available model names separated by comma (,)",isInput:true,isOutput:false},{name:"personaDescription",description:"Description of the persona",isInput:true,isOutput:false},{name:"modelRequirements",description:"Specific requirements for the model",isInput:false,isOutput:true}],tasks:[{taskType:"PROMPT_TASK",name:"make-model-requirements",title:"Make modelRequirements",content:"You are experienced AI engineer, you need to create virtual assistant.\nWrite\n\n## Example\n\n```json\n{\n\"modelName\": \"gpt-4o\",\n\"systemMessage\": \"You are experienced AI engineer and helpfull assistant.\",\n\"temperature\": 0.7\n}\n```\n\n## Instructions\n\n-   Your output format is JSON object\n-   Write just the JSON object, no other text should be present\n-   It contains the following keys:\n    -   `modelName`: The name of the model to use\n    -   `systemMessage`: The system message to provide context to the model\n    -   `temperature`: The sampling temperature to use\n\n### Key `modelName`\n\nPick from the following models:\n\n-   {availableModelNames}\n\n### Key `systemMessage`\n\nThe system message is used to communicate instructions or provide context to the model at the beginning of a conversation. It is displayed in a different format compared to user messages, helping the model understand its role in the conversation. The system message typically guides the model's behavior, sets the tone, or specifies desired output from the model. By utilizing the system message effectively, users can steer the model towards generating more accurate and relevant responses.\n\nFor example:\n\n> You are an experienced AI engineer and helpful assistant.\n\n> You are a friendly and knowledgeable chatbot.\n\n### Key `temperature`\n\nThe sampling temperature, between 0 and 1. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. If set to 0, the model will use log probability to automatically increase the temperature until certain thresholds are hit.\n\nYou can pick a value between 0 and 2. For example:\n\n-   `0.1`: Low temperature, extremely conservative and deterministic\n-   `0.5`: Medium temperature, balanced between conservative and creative\n-   `1.0`: High temperature, creative and bit random\n-   `1.5`: Very high temperature, extremely creative and often chaotic and unpredictable\n-   `2.0`: Maximum temperature, completely random and unpredictable, for some extreme creative use cases\n\n# The assistant\n\nTake this description of the persona:\n\n> {personaDescription}",resultingParameterName:"modelRequirements",format:"JSON",dependentParameterNames:["availableModelNames","personaDescription"]}],personas:[],preparations:[],knowledgeSources:[],knowledgePieces:[],sources:[{type:"BOOK",path:null,content:"# Prepare Persona\n\n-   PIPELINE URL `https://promptbook.studio/promptbook/prepare-persona.book`\n-   INPUT PARAMETER `{availableModelNames}` List of available model names separated by comma (,)\n-   INPUT PARAMETER `{personaDescription}` Description of the persona\n-   OUTPUT PARAMETER `{modelRequirements}` Specific requirements for the model\n\n## Make modelRequirements\n\n-   FORMAT JSON\n\n```markdown\nYou are experienced AI engineer, you need to create virtual assistant.\nWrite\n\n## Example\n\n\\`\\`\\`json\n{\n\"modelName\": \"gpt-4o\",\n\"systemMessage\": \"You are experienced AI engineer and helpfull assistant.\",\n\"temperature\": 0.7\n}\n\\`\\`\\`\n\n## Instructions\n\n-   Your output format is JSON object\n-   Write just the JSON object, no other text should be present\n-   It contains the following keys:\n    -   `modelName`: The name of the model to use\n    -   `systemMessage`: The system message to provide context to the model\n    -   `temperature`: The sampling temperature to use\n\n### Key `modelName`\n\nPick from the following models:\n\n-   {availableModelNames}\n\n### Key `systemMessage`\n\nThe system message is used to communicate instructions or provide context to the model at the beginning of a conversation. It is displayed in a different format compared to user messages, helping the model understand its role in the conversation. The system message typically guides the model's behavior, sets the tone, or specifies desired output from the model. By utilizing the system message effectively, users can steer the model towards generating more accurate and relevant responses.\n\nFor example:\n\n> You are an experienced AI engineer and helpful assistant.\n\n> You are a friendly and knowledgeable chatbot.\n\n### Key `temperature`\n\nThe sampling temperature, between 0 and 1. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. If set to 0, the model will use log probability to automatically increase the temperature until certain thresholds are hit.\n\nYou can pick a value between 0 and 2. For example:\n\n-   `0.1`: Low temperature, extremely conservative and deterministic\n-   `0.5`: Medium temperature, balanced between conservative and creative\n-   `1.0`: High temperature, creative and bit random\n-   `1.5`: Very high temperature, extremely creative and often chaotic and unpredictable\n-   `2.0`: Maximum temperature, completely random and unpredictable, for some extreme creative use cases\n\n# The assistant\n\nTake this description of the persona:\n\n> {personaDescription}\n```\n\n`-> {modelRequirements}`\n"}],sourceFile:"./books/prepare-persona.book"},{title:"Prepare Title",pipelineUrl:"https://promptbook.studio/promptbook/prepare-title.book",formfactorName:"GENERIC",parameters:[{name:"book",description:"The book to prepare the title for",isInput:true,isOutput:false},{name:"title",description:"Best title for the book",isInput:false,isOutput:true}],tasks:[{taskType:"PROMPT_TASK",name:"make-title",title:"Make title",content:"Make best title for given text which describes the workflow:\n\n## Rules\n\n-   Write just title, nothing else\n-   Title should be concise and clear - Write maximum ideally 2 words, maximum 5 words\n-   Title starts with emoticon\n-   Title should not mention the input and output of the workflow but the main purpose of the workflow\n    _For example, not \"✍ Convert Knowledge-piece to title\" but \"✍ Title\"_\n\n## The workflow\n\n> {book}",resultingParameterName:"title",expectations:{words:{min:1,max:8},lines:{min:1,max:1}},dependentParameterNames:["book"]}],personas:[],preparations:[],knowledgeSources:[],knowledgePieces:[],sources:[{type:"BOOK",path:null,content:"# Prepare Title\n\n-   PIPELINE URL `https://promptbook.studio/promptbook/prepare-title.book`\n-   INPUT PARAMETER `{book}` The book to prepare the title for\n-   OUTPUT PARAMETER `{title}` Best title for the book\n\n## Make title\n\n-   EXPECT MIN 1 Word\n-   EXPECT MAX 8 Words\n-   EXPECT EXACTLY 1 Line\n\n```markdown\nMake best title for given text which describes the workflow:\n\n## Rules\n\n-   Write just title, nothing else\n-   Title should be concise and clear - Write maximum ideally 2 words, maximum 5 words\n-   Title starts with emoticon\n-   Title should not mention the input and output of the workflow but the main purpose of the workflow\n    _For example, not \"✍ Convert Knowledge-piece to title\" but \"✍ Title\"_\n\n## The workflow\n\n> {book}\n```\n\n`-> {title}`\n"}],sourceFile:"./books/prepare-title.book"}];
+var PipelineCollection = [{title:"Prepare Knowledge from Markdown",pipelineUrl:"https://promptbook.studio/promptbook/prepare-knowledge-from-markdown.book",formfactorName:"GENERIC",parameters:[{name:"knowledgeContent",description:"Markdown document content",isInput:true,isOutput:false},{name:"knowledgePieces",description:"The knowledge JSON object",isInput:false,isOutput:true}],tasks:[{taskType:"PROMPT_TASK",name:"knowledge",title:"Knowledge",content:"You are experienced data researcher, extract the important knowledge from the document.\n\n# Rules\n\n-   Make pieces of information concise, clear, and easy to understand\n-   One piece of information should be approximately 1 paragraph\n-   Divide the paragraphs by markdown horizontal lines ---\n-   Omit irrelevant information\n-   Group redundant information\n-   Write just extracted information, nothing else\n\n# The document\n\nTake information from this document:\n\n> {knowledgeContent}",resultingParameterName:"knowledgePieces",dependentParameterNames:["knowledgeContent"]}],personas:[],preparations:[],knowledgeSources:[],knowledgePieces:[],sources:[{type:"BOOK",path:null,content:"# Prepare Knowledge from Markdown\n\n-   PIPELINE URL `https://promptbook.studio/promptbook/prepare-knowledge-from-markdown.book`\n-   INPUT PARAMETER `{knowledgeContent}` Markdown document content\n-   OUTPUT PARAMETER `{knowledgePieces}` The knowledge JSON object\n\n## Knowledge\n\n<!-- TODO: [🍆] -FORMAT JSON -->\n\n```markdown\nYou are experienced data researcher, extract the important knowledge from the document.\n\n# Rules\n\n-   Make pieces of information concise, clear, and easy to understand\n-   One piece of information should be approximately 1 paragraph\n-   Divide the paragraphs by markdown horizontal lines ---\n-   Omit irrelevant information\n-   Group redundant information\n-   Write just extracted information, nothing else\n\n# The document\n\nTake information from this document:\n\n> {knowledgeContent}\n```\n\n`-> {knowledgePieces}`\n"}],sourceFile:"./books/prepare-knowledge-from-markdown.book"},{title:"Prepare Keywords",pipelineUrl:"https://promptbook.studio/promptbook/prepare-knowledge-keywords.book",formfactorName:"GENERIC",parameters:[{name:"knowledgePieceContent",description:"The content",isInput:true,isOutput:false},{name:"keywords",description:"Keywords separated by comma",isInput:false,isOutput:true}],tasks:[{taskType:"PROMPT_TASK",name:"knowledge",title:"Knowledge",content:"You are experienced data researcher, detect the important keywords in the document.\n\n# Rules\n\n-   Write just keywords separated by comma\n\n# The document\n\nTake information from this document:\n\n> {knowledgePieceContent}",resultingParameterName:"keywords",dependentParameterNames:["knowledgePieceContent"]}],personas:[],preparations:[],knowledgeSources:[],knowledgePieces:[],sources:[{type:"BOOK",path:null,content:"# Prepare Keywords\n\n-   PIPELINE URL `https://promptbook.studio/promptbook/prepare-knowledge-keywords.book`\n-   INPUT PARAMETER `{knowledgePieceContent}` The content\n-   OUTPUT PARAMETER `{keywords}` Keywords separated by comma\n\n## Knowledge\n\n<!-- TODO: [🍆] -FORMAT JSON -->\n\n```markdown\nYou are experienced data researcher, detect the important keywords in the document.\n\n# Rules\n\n-   Write just keywords separated by comma\n\n# The document\n\nTake information from this document:\n\n> {knowledgePieceContent}\n```\n\n`-> {keywords}`\n"}],sourceFile:"./books/prepare-knowledge-keywords.book"},{title:"Prepare Knowledge-piece Title",pipelineUrl:"https://promptbook.studio/promptbook/prepare-knowledge-title.book",formfactorName:"GENERIC",parameters:[{name:"knowledgePieceContent",description:"The content",isInput:true,isOutput:false},{name:"title",description:"The title of the document",isInput:false,isOutput:true}],tasks:[{taskType:"PROMPT_TASK",name:"knowledge",title:"Knowledge",content:"You are experienced content creator, write best title for the document.\n\n# Rules\n\n-   Write just title, nothing else\n-   Write maximum 5 words for the title\n\n# The document\n\n> {knowledgePieceContent}",resultingParameterName:"title",expectations:{words:{min:1,max:8}},dependentParameterNames:["knowledgePieceContent"]}],personas:[],preparations:[],knowledgeSources:[],knowledgePieces:[],sources:[{type:"BOOK",path:null,content:"# Prepare Knowledge-piece Title\n\n-   PIPELINE URL `https://promptbook.studio/promptbook/prepare-knowledge-title.book`\n-   INPUT PARAMETER `{knowledgePieceContent}` The content\n-   OUTPUT PARAMETER `{title}` The title of the document\n\n## Knowledge\n\n-   EXPECT MIN 1 WORD\n-   EXPECT MAX 8 WORDS\n\n```markdown\nYou are experienced content creator, write best title for the document.\n\n# Rules\n\n-   Write just title, nothing else\n-   Write maximum 5 words for the title\n\n# The document\n\n> {knowledgePieceContent}\n```\n\n`-> {title}`\n"}],sourceFile:"./books/prepare-knowledge-title.book"},{title:"Prepare Persona",pipelineUrl:"https://promptbook.studio/promptbook/prepare-persona.book",formfactorName:"GENERIC",parameters:[{name:"availableModels",description:"List of available model names together with their descriptions as JSON",isInput:true,isOutput:false},{name:"personaDescription",description:"Description of the persona",isInput:true,isOutput:false},{name:"modelsRequirements",description:"Specific requirements for the model",isInput:false,isOutput:true}],tasks:[{taskType:"PROMPT_TASK",name:"make-model-requirements",title:"Make modelRequirements",content:"You are an experienced AI engineer, you need to find the best models for virtual assistants:\n\n## Example\n\n```json\n[\n  {\n  \"modelName\": \"gpt-4o\",\n  \"systemMessage\": \"You are experienced AI engineer and helpfull assistant.\",\n  \"temperature\": 0.7\n  },\n  {\n    \"modelName\": \"claude-3-5-sonnet\",\n    \"systemMessage\": \"You are a friendly and knowledgeable chatbot.\",\n    \"temperature\": 0.5\n  }\n]\n```\n\n## Instructions\n\n-   Your output format is JSON array\n-   Sort best-fitting models first\n-   Omit any models that are not suitable\n-   Write just the JSON, no other text should be present\n-   Array contain items with following keys:\n    -   `modelName`: The name of the model to use\n    -   `systemMessage`: The system message to provide context to the model\n    -   `temperature`: The sampling temperature to use\n\n### Key `modelName`\n\nHere are the available models:\n\n```json\n{availableModels}\n```\n\n### Key `systemMessage`\n\nThe system message is used to communicate instructions or provide context to the model at the beginning of a conversation. It is displayed in a different format compared to user messages, helping the model understand its role in the conversation. The system message typically guides the model's behavior, sets the tone, or specifies desired output from the model. By utilizing the system message effectively, users can steer the model towards generating more accurate and relevant responses.\n\nFor example:\n\n> You are an experienced AI engineer and helpful assistant.\n\n> You are a friendly and knowledgeable chatbot.\n\n### Key `temperature`\n\nThe sampling temperature, between 0 and 1. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. If set to 0, the model will use log probability to automatically increase the temperature until certain thresholds are hit.\n\nYou can pick a value between 0 and 2. For example:\n\n-   `0.1`: Low temperature, extremely conservative and deterministic\n-   `0.5`: Medium temperature, balanced between conservative and creative\n-   `1.0`: High temperature, creative and bit random\n-   `1.5`: Very high temperature, extremely creative and often chaotic and unpredictable\n-   `2.0`: Maximum temperature, completely random and unpredictable, for some extreme creative use cases\n\n# The assistant\n\nTake this description of the persona:\n\n> {personaDescription}",resultingParameterName:"modelsRequirements",format:"JSON",dependentParameterNames:["availableModels","personaDescription"]}],personas:[],preparations:[],knowledgeSources:[],knowledgePieces:[],sources:[{type:"BOOK",path:null,content:"# Prepare Persona\n\n-   PIPELINE URL `https://promptbook.studio/promptbook/prepare-persona.book`\n-   INPUT PARAMETER `{availableModels}` List of available model names together with their descriptions as JSON\n-   INPUT PARAMETER `{personaDescription}` Description of the persona\n-   OUTPUT PARAMETER `{modelsRequirements}` Specific requirements for the model\n\n## Make modelRequirements\n\n-   FORMAT JSON\n\n```markdown\nYou are an experienced AI engineer, you need to find the best models for virtual assistants:\n\n## Example\n\n\\`\\`\\`json\n[\n  {\n  \"modelName\": \"gpt-4o\",\n  \"systemMessage\": \"You are experienced AI engineer and helpfull assistant.\",\n  \"temperature\": 0.7\n  },\n  {\n    \"modelName\": \"claude-3-5-sonnet\",\n    \"systemMessage\": \"You are a friendly and knowledgeable chatbot.\",\n    \"temperature\": 0.5\n  }\n]\n\\`\\`\\`\n\n## Instructions\n\n-   Your output format is JSON array\n-   Sort best-fitting models first\n-   Omit any models that are not suitable\n-   Write just the JSON, no other text should be present\n-   Array contain items with following keys:\n    -   `modelName`: The name of the model to use\n    -   `systemMessage`: The system message to provide context to the model\n    -   `temperature`: The sampling temperature to use\n\n### Key `modelName`\n\nHere are the available models:\n\n\\`\\`\\`json\n{availableModels}\n\\`\\`\\`\n\n### Key `systemMessage`\n\nThe system message is used to communicate instructions or provide context to the model at the beginning of a conversation. It is displayed in a different format compared to user messages, helping the model understand its role in the conversation. The system message typically guides the model's behavior, sets the tone, or specifies desired output from the model. By utilizing the system message effectively, users can steer the model towards generating more accurate and relevant responses.\n\nFor example:\n\n> You are an experienced AI engineer and helpful assistant.\n\n> You are a friendly and knowledgeable chatbot.\n\n### Key `temperature`\n\nThe sampling temperature, between 0 and 1. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. If set to 0, the model will use log probability to automatically increase the temperature until certain thresholds are hit.\n\nYou can pick a value between 0 and 2. For example:\n\n-   `0.1`: Low temperature, extremely conservative and deterministic\n-   `0.5`: Medium temperature, balanced between conservative and creative\n-   `1.0`: High temperature, creative and bit random\n-   `1.5`: Very high temperature, extremely creative and often chaotic and unpredictable\n-   `2.0`: Maximum temperature, completely random and unpredictable, for some extreme creative use cases\n\n# The assistant\n\nTake this description of the persona:\n\n> {personaDescription}\n```\n\n`-> {modelsRequirements}`\n"}],sourceFile:"./books/prepare-persona.book"},{title:"Prepare Title",pipelineUrl:"https://promptbook.studio/promptbook/prepare-title.book",formfactorName:"GENERIC",parameters:[{name:"book",description:"The book to prepare the title for",isInput:true,isOutput:false},{name:"title",description:"Best title for the book",isInput:false,isOutput:true}],tasks:[{taskType:"PROMPT_TASK",name:"make-title",title:"Make title",content:"Make best title for given text which describes the workflow:\n\n## Rules\n\n-   Write just title, nothing else\n-   Title should be concise and clear - Write maximum ideally 2 words, maximum 5 words\n-   Title starts with emoticon\n-   Title should not mention the input and output of the workflow but the main purpose of the workflow\n    _For example, not \"✍ Convert Knowledge-piece to title\" but \"✍ Title\"_\n\n## The workflow\n\n> {book}",resultingParameterName:"title",expectations:{words:{min:1,max:8},lines:{min:1,max:1}},dependentParameterNames:["book"]}],personas:[],preparations:[],knowledgeSources:[],knowledgePieces:[],sources:[{type:"BOOK",path:null,content:"# Prepare Title\n\n-   PIPELINE URL `https://promptbook.studio/promptbook/prepare-title.book`\n-   INPUT PARAMETER `{book}` The book to prepare the title for\n-   OUTPUT PARAMETER `{title}` Best title for the book\n\n## Make title\n\n-   EXPECT MIN 1 Word\n-   EXPECT MAX 8 Words\n-   EXPECT EXACTLY 1 Line\n\n```markdown\nMake best title for given text which describes the workflow:\n\n## Rules\n\n-   Write just title, nothing else\n-   Title should be concise and clear - Write maximum ideally 2 words, maximum 5 words\n-   Title starts with emoticon\n-   Title should not mention the input and output of the workflow but the main purpose of the workflow\n    _For example, not \"✍ Convert Knowledge-piece to title\" but \"✍ Title\"_\n\n## The workflow\n\n> {book}\n```\n\n`-> {title}`\n"}],sourceFile:"./books/prepare-title.book"}];
 
 /**
  * This error type indicates that some tools are missing for pipeline execution or preparation
@@ -1854,7 +1854,7 @@ function isPipelinePrepared(pipeline) {
     if (pipeline.title === undefined || pipeline.title === '' || pipeline.title === DEFAULT_BOOK_TITLE) {
         return false;
     }
-    if (!pipeline.personas.every((persona) => persona.modelRequirements !== undefined)) {
+    if (!pipeline.personas.every((persona) => persona.modelsRequirements !== undefined)) {
         return false;
     }
     if (!pipeline.knowledgeSources.every((knowledgeSource) => knowledgeSource.preparationIds !== undefined)) {
@@ -1893,6 +1893,45 @@ function $randomToken(randomness) {
  * TODO: Maybe use nanoid instead https://github.com/ai/nanoid
  */
 
+/**
+ * Converts a JavaScript Object Notation (JSON) string into an object.
+ *
+ * Note: This is wrapper around `JSON.parse()` with better error and type handling
+ *
+ * @public exported from `@promptbook/utils`
+ */
+function jsonParse(value) {
+    if (value === undefined) {
+        throw new Error(`Can not parse JSON from undefined value.`);
+    }
+    else if (typeof value !== 'string') {
+        console.error('Can not parse JSON from non-string value.', { text: value });
+        throw new Error(spaceTrim(`
+                    Can not parse JSON from non-string value.
+
+                    The value type: ${typeof value}
+                    See more in console.
+                `));
+    }
+    try {
+        return JSON.parse(value);
+    }
+    catch (error) {
+        if (!(error instanceof Error)) {
+            throw error;
+        }
+        throw new Error(spaceTrim((block) => `
+                    ${block(error.message)}
+            
+                    The JSON text:
+                    ${block(value)}
+                `));
+    }
+}
+/**
+ * TODO: !!!! Use in Promptbook.studio
+ */
+
 /**
  * Recursively converts JSON strings to JSON objects
 
@@ -1911,7 +1950,7 @@ function jsonStringsToJsons(object) {
     const newObject = { ...object };
     for (const [key, value] of Object.entries(object)) {
         if (typeof value === 'string' && isValidJsonString(value)) {
-            newObject[key] = JSON.parse(value);
+            newObject[key] = jsonParse(value);
         }
         else {
             newObject[key] = jsonStringsToJsons(value);
@@ -2733,6 +2772,24 @@ function isValidCsvString(value) {
     }
 }
 
+/**
+ * Converts a CSV string into an object
+ *
+ * Note: This is wrapper around `papaparse.parse()` with better autohealing
+ *
+ * @private - for now until `@promptbook/csv` is released
+ */
+function csvParse(value /* <- TODO: string_csv */, settings, schema /* <- TODO: Make CSV Schemas */) {
+    settings = { ...settings, ...MANDATORY_CSV_SETTINGS };
+    // Note: Autoheal invalid '\n' characters
+    if (settings.newline && !settings.newline.includes('\r') && value.includes('\r')) {
+        console.warn('CSV string contains carriage return characters, but in the CSV settings the `newline` setting does not include them. Autohealing the CSV string.');
+        value = value.replace(/\r\n/g, '\n').replace(/\r/g, '\n');
+    }
+    const csv = parse(value, settings);
+    return csv;
+}
+
 /**
  * Definition for CSV spreadsheet
  *
@@ -2755,8 +2812,7 @@ const CsvFormatDefinition = {
         {
             subvalueName: 'ROW',
             async mapValues(value, outputParameterName, settings, mapCallback) {
-                // TODO: [👨🏾‍🤝‍👨🏼] DRY csv parsing
-                const csv = parse(value, { ...settings, ...MANDATORY_CSV_SETTINGS });
+                const csv = csvParse(value, settings);
                 if (csv.errors.length !== 0) {
                     throw new CsvFormatError(spaceTrim((block) => `
                                 CSV parsing error
@@ -2786,8 +2842,7 @@ const CsvFormatDefinition = {
         {
             subvalueName: 'CELL',
             async mapValues(value, outputParameterName, settings, mapCallback) {
-                // TODO: [👨🏾‍🤝‍👨🏼] DRY csv parsing
-                const csv = parse(value, { ...settings, ...MANDATORY_CSV_SETTINGS });
+                const csv = csvParse(value, settings);
                 if (csv.errors.length !== 0) {
                     throw new CsvFormatError(spaceTrim((block) => `
                                 CSV parsing error
@@ -4327,13 +4382,79 @@ async function getExamplesForTask(task) {
 /**
  * @@@
  *
+ * Here is the place where RAG (retrieval-augmented generation) happens
+ *
  * @private internal utility of `createPipelineExecutor`
  */
 async function getKnowledgeForTask(options) {
-    const { preparedPipeline, task } = options;
-    return preparedPipeline.knowledgePieces.map(({ content }) => `- ${content}`).join('\n');
+    const { tools, preparedPipeline, task } = options;
+    const firstKnowlegePiece = preparedPipeline.knowledgePieces[0];
+    const firstKnowlegeIndex = firstKnowlegePiece === null || firstKnowlegePiece === void 0 ? void 0 : firstKnowlegePiece.index[0];
+    // <- TODO: Do not use just first knowledge piece and first index to determine embedding model, use also keyword search
+    if (firstKnowlegePiece === undefined || firstKnowlegeIndex === undefined) {
+        return 'No knowledge pieces found';
+    }
+    // TODO: [🚐] Make arrayable LLMs -> single LLM DRY
+    const _llms = arrayableToArray(tools.llm);
+    const llmTools = _llms.length === 1 ? _llms[0] : joinLlmExecutionTools(..._llms);
+    const taskEmbeddingPrompt = {
+        title: 'Knowledge Search',
+        modelRequirements: {
+            modelVariant: 'EMBEDDING',
+            modelName: firstKnowlegeIndex.modelName,
+        },
+        content: task.content,
+        parameters: {
+        /* !!!!!!!! */
+        },
+    };
+    const taskEmbeddingResult = await llmTools.callEmbeddingModel(taskEmbeddingPrompt);
+    const knowledgePiecesWithRelevance = preparedPipeline.knowledgePieces.map((knowledgePiece) => {
+        const { index } = knowledgePiece;
+        const knowledgePieceIndex = index.find((i) => i.modelName === firstKnowlegeIndex.modelName);
+        // <- TODO: Do not use just first knowledge piece and first index to determine embedding model
+        if (knowledgePieceIndex === undefined) {
+            return {
+                content: knowledgePiece.content,
+                relevance: 0,
+            };
+        }
+        const relevance = computeCosineSimilarity(knowledgePieceIndex.position, taskEmbeddingResult.content);
+        return {
+            content: knowledgePiece.content,
+            relevance,
+        };
+    });
+    const knowledgePiecesSorted = knowledgePiecesWithRelevance.sort((a, b) => a.relevance - b.relevance);
+    const knowledgePiecesLimited = knowledgePiecesSorted.slice(0, 5);
+    console.log('!!! Embedding', {
+        task,
+        taskEmbeddingPrompt,
+        taskEmbeddingResult,
+        firstKnowlegePiece,
+        firstKnowlegeIndex,
+        knowledgePiecesWithRelevance,
+        knowledgePiecesSorted,
+        knowledgePiecesLimited,
+    });
+    return knowledgePiecesLimited.map(({ content }) => `- ${content}`).join('\n');
     //                                                      <- TODO: [🧠] Some smart aggregation of knowledge pieces, single-line vs multi-line vs mixed
 }
+// TODO: !!!!!! Annotate + to new file
+function computeCosineSimilarity(embeddingVector1, embeddingVector2) {
+    if (embeddingVector1.length !== embeddingVector2.length) {
+        throw new TypeError('Embedding vectors must have the same length');
+    }
+    const dotProduct = embeddingVector1.reduce((sum, value, index) => sum + value * embeddingVector2[index], 0);
+    const magnitude1 = Math.sqrt(embeddingVector1.reduce((sum, value) => sum + value * value, 0));
+    const magnitude2 = Math.sqrt(embeddingVector2.reduce((sum, value) => sum + value * value, 0));
+    return 1 - dotProduct / (magnitude1 * magnitude2);
+}
+/**
+ * TODO: !!!! Verify if this is working
+ * TODO: [♨] Implement Better - use keyword search
+ * TODO: [♨] Examples of values
+ */
 
 /**
  * @@@
@@ -4341,9 +4462,9 @@ async function getKnowledgeForTask(options) {
  * @private internal utility of `createPipelineExecutor`
  */
 async function getReservedParametersForTask(options) {
-    const { preparedPipeline, task, pipelineIdentification } = options;
+    const { tools, preparedPipeline, task, pipelineIdentification } = options;
     const context = await getContextForTask(); // <- [🏍]
-    const knowledge = await getKnowledgeForTask({ preparedPipeline, task });
+    const knowledge = await getKnowledgeForTask({ tools, preparedPipeline, task });
     const examples = await getExamplesForTask();
     const currentDate = new Date().toISOString(); // <- TODO: [🧠][💩] Better
     const modelName = RESERVED_PARAMETER_MISSING_VALUE;
@@ -4405,6 +4526,7 @@ async function executeTask(options) {
     }
     const definedParameters = Object.freeze({
         ...(await getReservedParametersForTask({
+            tools,
             preparedPipeline,
             task: currentTask,
             pipelineIdentification,
@@ -4961,27 +5083,48 @@ async function preparePersona(personaDescription, tools, options) {
         pipeline: await collection.getPipelineByUrl('https://promptbook.studio/promptbook/prepare-persona.book'),
         tools,
     });
-    // TODO: [🚐] Make arrayable LLMs -> single LLM DRY
     const _llms = arrayableToArray(tools.llm);
     const llmTools = _llms.length === 1 ? _llms[0] : joinLlmExecutionTools(..._llms);
-    const availableModels = await llmTools.listModels();
-    const availableModelNames = availableModels
+    const availableModels = (await llmTools.listModels())
         .filter(({ modelVariant }) => modelVariant === 'CHAT')
-        .map(({ modelName }) => modelName)
-        .join(',');
-    const result = await preparePersonaExecutor({ availableModelNames, personaDescription }).asPromise();
+        .map(({ modelName, modelDescription }) => ({
+        modelName,
+        modelDescription,
+        // <- Note: `modelTitle` and `modelVariant` is not relevant for this task
+    }));
+    const result = await preparePersonaExecutor({
+        availableModels /* <- Note: Passing as JSON */,
+        personaDescription,
+    }).asPromise();
     const { outputParameters } = result;
-    const { modelRequirements: modelRequirementsRaw } = outputParameters;
-    const modelRequirements = JSON.parse(modelRequirementsRaw);
+    const { modelsRequirements: modelsRequirementsJson } = outputParameters;
+    let modelsRequirementsUnchecked = jsonParse(modelsRequirementsJson);
     if (isVerbose) {
-        console.info(`PERSONA ${personaDescription}`, modelRequirements);
+        console.info(`PERSONA ${personaDescription}`, modelsRequirementsUnchecked);
     }
-    const { modelName, systemMessage, temperature } = modelRequirements;
-    return {
+    if (!Array.isArray(modelsRequirementsUnchecked)) {
+        // <- TODO: Book should have syntax and system to enforce shape of JSON
+        modelsRequirementsUnchecked = [modelsRequirementsUnchecked];
+        /*
+        throw new UnexpectedError(
+            spaceTrim(
+                (block) => `
+                    Invalid \`modelsRequirements\`:
+
+                    \`\`\`json
+                    ${block(JSON.stringify(modelsRequirementsUnchecked, null, 4))}
+                    \`\`\`
+                `,
+            ),
+        );
+        */
+    }
+    const modelsRequirements = modelsRequirementsUnchecked.map((modelRequirements) => ({
         modelVariant: 'CHAT',
-        modelName,
-        systemMessage,
-        temperature,
+        ...modelRequirements,
+    }));
+    return {
+        modelsRequirements,
     };
 }
 /**
@@ -5550,7 +5693,7 @@ async function makeKnowledgeSourceHandler(knowledgeSource, tools, options) {
                 > },
             */
             async asJson() {
-                return JSON.parse(await tools.fs.readFile(filename, 'utf-8'));
+                return jsonParse(await tools.fs.readFile(filename, 'utf-8'));
             },
             async asText() {
                 return await tools.fs.readFile(filename, 'utf-8');
@@ -5808,14 +5951,14 @@ async function preparePipeline(pipeline, tools, options) {
     // TODO: [🖌][🧠] Implement some `mapAsync` function
     const preparedPersonas = new Array(personas.length);
     await forEachAsync(personas, { maxParallelCount /* <- TODO: [🪂] When there are subtasks, this maximul limit can be broken */ }, async (persona, index) => {
-        const modelRequirements = await preparePersona(persona.description, { ...tools, llm: llmToolsWithUsage }, {
+        const { modelsRequirements } = await preparePersona(persona.description, { ...tools, llm: llmToolsWithUsage }, {
             rootDirname,
             maxParallelCount /* <- TODO:  [🪂] */,
             isVerbose,
         });
         const preparedPersona = {
             ...persona,
-            modelRequirements,
+            modelsRequirements,
             preparationIds: [/* TODO: [🧊] -> */ currentPreparation.id],
             // <- TODO: [🍙] Make some standard order of json properties
         };
@@ -9904,6 +10047,81 @@ const BoilerplateFormfactorDefinition = {
     },
 };
 
+/**
+ * Creates a wrapper around LlmExecutionTools that only exposes models matching the filter function
+ *
+ * @param llmTools The original LLM execution tools to wrap
+ * @param modelFilter Function that determines whether a model should be included
+ * @returns A new LlmExecutionTools instance with filtered models
+ *
+ * @public exported from `@promptbook/core`
+ */
+function filterModels(llmTools, modelFilter) {
+    const filteredTools = {
+        // Keep all properties from the original llmTools
+        ...llmTools,
+        get description() {
+            return `${llmTools.description} (filtered)`;
+        },
+        // Override listModels to filter the models
+        async listModels() {
+            const originalModels = await llmTools.listModels();
+            // Handle both synchronous and Promise return types
+            if (originalModels instanceof Promise) {
+                return originalModels.then((models) => models.filter(modelFilter));
+            }
+            else {
+                return originalModels.filter(modelFilter);
+            }
+        },
+    };
+    // Helper function to validate if a model is allowed
+    async function isModelAllowed(modelName) {
+        const models = await filteredTools.listModels();
+        return models.some((model) => model.modelName === modelName);
+    }
+    // Override callChatModel if it exists in the original tools
+    if (llmTools.callChatModel) {
+        filteredTools.callChatModel = async (prompt) => {
+            var _a;
+            const modelName = (_a = prompt.modelRequirements) === null || _a === void 0 ? void 0 : _a.modelName;
+            // If a specific model is requested, check if it's allowed
+            if (modelName && !(await isModelAllowed(modelName))) {
+                throw new PipelineExecutionError(`Model ${modelName} is not allowed by the filter for chat calls`);
+            }
+            return llmTools.callChatModel(prompt);
+        };
+    }
+    // Override callCompletionModel if it exists in the original tools
+    if (llmTools.callCompletionModel) {
+        filteredTools.callCompletionModel = async (prompt) => {
+            var _a;
+            const modelName = (_a = prompt.modelRequirements) === null || _a === void 0 ? void 0 : _a.modelName;
+            // If a specific model is requested, check if it's allowed
+            if (modelName && !(await isModelAllowed(modelName))) {
+                throw new PipelineExecutionError(`Model ${modelName} is not allowed by the filter for completion calls`);
+            }
+            return llmTools.callCompletionModel(prompt);
+        };
+    }
+    // Override callEmbeddingModel if it exists in the original tools
+    if (llmTools.callEmbeddingModel) {
+        filteredTools.callEmbeddingModel = async (prompt) => {
+            var _a;
+            const modelName = (_a = prompt.modelRequirements) === null || _a === void 0 ? void 0 : _a.modelName;
+            // If a specific model is requested, check if it's allowed
+            if (modelName && !(await isModelAllowed(modelName))) {
+                throw new PipelineExecutionError(`Model ${modelName} is not allowed by the filter for embedding calls`);
+            }
+            return llmTools.callEmbeddingModel(prompt);
+        };
+    }
+    return filteredTools;
+}
+/**
+ * TODO: !!! [models] Test that this is working
+ */
+
 /**
  * @@@
  *
@@ -10121,6 +10339,48 @@ function createLlmToolsFromConfiguration(configuration, options = {}) {
  * TODO: [®] DRY Register logic
  */
 
+/**
+ * How is the model provider trusted?
+ *
+ * @public exported from `@promptbook/core`
+ */
+const MODEL_TRUST_LEVEL = {
+    FULL: `Model is running on the local machine, training data and model weights are known, data are ethically sourced`,
+    OPEN: `Model is open source, training data and model weights are known`,
+    PARTIALLY_OPEN: `Model is open source, but training data and model weights are not (fully) known`,
+    CLOSED_LOCAL: `Model can be run locally, but it is not open source`,
+    CLOSED_FREE: `Model is behind API gateway but free to use`,
+    CLOSED_BUSINESS: `Model is behind API gateway and paid but has good SLA, TOS, privacy policy and in general is a good to use in business applications`,
+    CLOSED: `Model is behind API gateway and paid`,
+    UNTRUSTED: `Model has questions about the training data and ethics, but it is not known if it is a problem or not`,
+    VURNABLE: `Model has some known serious vulnerabilities, leaks, ethical problems, etc.`,
+};
+// <- TODO: Maybe do better levels of trust
+/**
+ * How is the model provider important?
+ *
+ * @public exported from `@promptbook/core`
+ */
+const MODEL_ORDER = {
+    /**
+     * Top-tier models, e.g. OpenAI, Anthropic,...
+     */
+    TOP_TIER: 333,
+    /**
+     * Mid-tier models, e.g. Llama, Mistral, etc.
+     */
+    NORMAL: 100,
+    /**
+     * Low-tier models, e.g. Phi, Tiny, etc.
+     */
+    LOW_TIER: 0,
+};
+/**
+ * TODO: Add configuration schema and maybe some documentation link
+ * TODO: Maybe constrain LlmToolsConfiguration[number] by generic to ensure that `createConfigurationFromEnv` and `getBoilerplateConfiguration` always create same `packageName` and `className`
+ * TODO: [®] DRY Register logic
+ */
+
 /**
  * Stores data in memory (HEAP)
  *
@@ -10320,9 +10580,11 @@ const _AnthropicClaudeMetadataRegistration = $llmToolsMetadataRegister.register(
     packageName: '@promptbook/anthropic-claude',
     className: 'AnthropicClaudeExecutionTools',
     envVariables: ['ANTHROPIC_CLAUDE_API_KEY'],
+    trustLevel: 'CLOSED',
+    order: MODEL_ORDER.TOP_TIER,
     getBoilerplateConfiguration() {
         return {
-            title: 'Anthropic Claude (boilerplate)',
+            title: 'Anthropic Claude',
             packageName: '@promptbook/anthropic-claude',
             className: 'AnthropicClaudeExecutionTools',
             options: {
@@ -10365,9 +10627,11 @@ const _AzureOpenAiMetadataRegistration = $llmToolsMetadataRegister.register({
     packageName: '@promptbook/azure-openai',
     className: 'AzureOpenAiExecutionTools',
     envVariables: ['AZUREOPENAI_RESOURCE_NAME', 'AZUREOPENAI_DEPLOYMENT_NAME', 'AZUREOPENAI_API_KEY'],
+    trustLevel: 'CLOSED_BUSINESS',
+    order: MODEL_ORDER.NORMAL,
     getBoilerplateConfiguration() {
         return {
-            title: 'Azure Open AI (boilerplate)',
+            title: 'Azure Open AI',
             packageName: '@promptbook/azure-openai',
             className: 'AzureOpenAiExecutionTools',
             options: {
@@ -10451,9 +10715,11 @@ const _DeepseekMetadataRegistration = $llmToolsMetadataRegister.register({
     packageName: '@promptbook/deepseek',
     className: 'DeepseekExecutionTools',
     envVariables: ['DEEPSEEK_GENERATIVE_AI_API_KEY'],
+    trustLevel: 'UNTRUSTED',
+    order: MODEL_ORDER.NORMAL,
     getBoilerplateConfiguration() {
         return {
-            title: 'Deepseek (boilerplate)',
+            title: 'Deepseek',
             packageName: '@promptbook/deepseek',
             className: 'DeepseekExecutionTools',
             options: {
@@ -10500,9 +10766,11 @@ const _GoogleMetadataRegistration = $llmToolsMetadataRegister.register({
     packageName: '@promptbook/google',
     className: 'GoogleExecutionTools',
     envVariables: ['GOOGLE_GENERATIVE_AI_API_KEY'],
+    trustLevel: 'CLOSED',
+    order: MODEL_ORDER.NORMAL,
     getBoilerplateConfiguration() {
         return {
-            title: 'Google Gemini (boilerplate)',
+            title: 'Google Gemini',
             packageName: '@promptbook/google',
             className: 'GoogleExecutionTools',
             options: {
@@ -10549,9 +10817,11 @@ const _OpenAiMetadataRegistration = $llmToolsMetadataRegister.register({
     packageName: '@promptbook/openai',
     className: 'OpenAiExecutionTools',
     envVariables: ['OPENAI_API_KEY'],
+    trustLevel: 'CLOSED',
+    order: MODEL_ORDER.TOP_TIER,
     getBoilerplateConfiguration() {
         return {
-            title: 'Open AI (boilerplate)',
+            title: 'Open AI',
             packageName: '@promptbook/openai',
             className: 'OpenAiExecutionTools',
             options: {
@@ -10589,9 +10859,11 @@ const _OpenAiAssistantMetadataRegistration = $llmToolsMetadataRegister.register(
     className: 'OpenAiAssistantExecutionTools',
     envVariables: null,
     //            <- TODO: ['OPENAI_API_KEY', 'OPENAI_ASSISTANT_ID']
+    trustLevel: 'CLOSED',
+    order: MODEL_ORDER.NORMAL,
     getBoilerplateConfiguration() {
         return {
-            title: 'Open AI Assistant (boilerplate)',
+            title: 'Open AI Assistant',
             packageName: '@promptbook/openai',
             className: 'OpenAiAssistantExecutionTools',
             options: {
@@ -11055,5 +11327,5 @@ class PrefixStorage {
     }
 }
 
-export { $llmToolsMetadataRegister, $llmToolsRegister, $scrapersMetadataRegister, $scrapersRegister, ADMIN_EMAIL, ADMIN_GITHUB_NAME, AbstractFormatError, AuthenticationError, BOOK_LANGUAGE_VERSION, BlackholeStorage, BoilerplateError, BoilerplateFormfactorDefinition, CLAIM, CLI_APP_ID, CallbackInterfaceTools, ChatbotFormfactorDefinition, CollectionError, CsvFormatDefinition, CsvFormatError, DEFAULT_BOOKS_DIRNAME, DEFAULT_BOOK_OUTPUT_PARAMETER_NAME, DEFAULT_BOOK_TITLE, DEFAULT_CSV_SETTINGS, DEFAULT_DOWNLOAD_CACHE_DIRNAME, DEFAULT_EXECUTION_CACHE_DIRNAME, DEFAULT_GET_PIPELINE_COLLECTION_FUNCTION_NAME, DEFAULT_INTERMEDIATE_FILES_STRATEGY, DEFAULT_IS_AUTO_INSTALLED, DEFAULT_IS_VERBOSE, DEFAULT_MAX_EXECUTION_ATTEMPTS, DEFAULT_MAX_FILE_SIZE, DEFAULT_MAX_KNOWLEDGE_SOURCES_SCRAPING_DEPTH, DEFAULT_MAX_KNOWLEDGE_SOURCES_SCRAPING_TOTAL, DEFAULT_MAX_PARALLEL_COUNT, DEFAULT_PIPELINE_COLLECTION_BASE_FILENAME, DEFAULT_PROMPT_TASK_TITLE, DEFAULT_REMOTE_SERVER_URL, DEFAULT_SCRAPE_CACHE_DIRNAME, DEFAULT_TASK_TITLE, EXPECTATION_UNITS, EnvironmentMismatchError, ExecutionReportStringOptionsDefaults, ExpectError, FORMFACTOR_DEFINITIONS, GENERIC_PIPELINE_INTERFACE, GeneratorFormfactorDefinition, GenericFormfactorDefinition, ImageGeneratorFormfactorDefinition, KnowledgeScrapeError, LimitReachedError, MANDATORY_CSV_SETTINGS, MAX_FILENAME_LENGTH, MODEL_VARIANTS, MatcherFormfactorDefinition, MemoryStorage, MissingToolsError, MultipleLlmExecutionTools, NAME, NonTaskSectionTypes, NotFoundError, NotYetImplementedError, ORDER_OF_PIPELINE_JSON, PLAYGROUND_APP_ID, PROMPTBOOK_ENGINE_VERSION, PROMPTBOOK_ERRORS, ParseError, PipelineExecutionError, PipelineLogicError, PipelineUrlError, PrefixStorage, PromptbookFetchError, REMOTE_SERVER_URLS, RESERVED_PARAMETER_NAMES, SET_IS_VERBOSE, SectionTypes, SheetsFormfactorDefinition, TaskTypes, TextFormatDefinition, TranslatorFormfactorDefinition, UNCERTAIN_USAGE, UNCERTAIN_ZERO_VALUE, UnexpectedError, WrappedError, ZERO_USAGE, ZERO_VALUE, _AnthropicClaudeMetadataRegistration, _AzureOpenAiMetadataRegistration, _BoilerplateScraperMetadataRegistration, _DeepseekMetadataRegistration, _DocumentScraperMetadataRegistration, _GoogleMetadataRegistration, _LegacyDocumentScraperMetadataRegistration, _MarkdownScraperMetadataRegistration, _MarkitdownScraperMetadataRegistration, _OpenAiAssistantMetadataRegistration, _OpenAiMetadataRegistration, _PdfScraperMetadataRegistration, _WebsiteScraperMetadataRegistration, addUsage, book, cacheLlmTools, collectionToJson, compilePipeline, countUsage, createCollectionFromJson, createCollectionFromPromise, createCollectionFromUrl, createLlmToolsFromConfiguration, createPipelineExecutor, createSubcollection, embeddingVectorToString, executionReportJsonToString, extractParameterNamesFromTask, getPipelineInterface, identificationToPromptbookToken, isPassingExpectations, isPipelineImplementingInterface, isPipelineInterfacesEqual, isPipelinePrepared, isValidPipelineString, joinLlmExecutionTools, limitTotalUsage, makeKnowledgeSourceHandler, parsePipeline, pipelineJsonToString, prepareKnowledgePieces, preparePersona, preparePipeline, prepareTasks, prettifyPipelineString, promptbookFetch, promptbookTokenToIdentification, unpreparePipeline, usageToHuman, usageToWorktime, validatePipeline, validatePipelineString };
+export { $llmToolsMetadataRegister, $llmToolsRegister, $scrapersMetadataRegister, $scrapersRegister, ADMIN_EMAIL, ADMIN_GITHUB_NAME, AbstractFormatError, AuthenticationError, BOOK_LANGUAGE_VERSION, BlackholeStorage, BoilerplateError, BoilerplateFormfactorDefinition, CLAIM, CLI_APP_ID, CallbackInterfaceTools, ChatbotFormfactorDefinition, CollectionError, CsvFormatDefinition, CsvFormatError, DEFAULT_BOOKS_DIRNAME, DEFAULT_BOOK_OUTPUT_PARAMETER_NAME, DEFAULT_BOOK_TITLE, DEFAULT_CSV_SETTINGS, DEFAULT_DOWNLOAD_CACHE_DIRNAME, DEFAULT_EXECUTION_CACHE_DIRNAME, DEFAULT_GET_PIPELINE_COLLECTION_FUNCTION_NAME, DEFAULT_INTERMEDIATE_FILES_STRATEGY, DEFAULT_IS_AUTO_INSTALLED, DEFAULT_IS_VERBOSE, DEFAULT_MAX_EXECUTION_ATTEMPTS, DEFAULT_MAX_FILE_SIZE, DEFAULT_MAX_KNOWLEDGE_SOURCES_SCRAPING_DEPTH, DEFAULT_MAX_KNOWLEDGE_SOURCES_SCRAPING_TOTAL, DEFAULT_MAX_PARALLEL_COUNT, DEFAULT_PIPELINE_COLLECTION_BASE_FILENAME, DEFAULT_PROMPT_TASK_TITLE, DEFAULT_REMOTE_SERVER_URL, DEFAULT_SCRAPE_CACHE_DIRNAME, DEFAULT_TASK_TITLE, EXPECTATION_UNITS, EnvironmentMismatchError, ExecutionReportStringOptionsDefaults, ExpectError, FORMFACTOR_DEFINITIONS, GENERIC_PIPELINE_INTERFACE, GeneratorFormfactorDefinition, GenericFormfactorDefinition, ImageGeneratorFormfactorDefinition, KnowledgeScrapeError, LimitReachedError, MANDATORY_CSV_SETTINGS, MAX_FILENAME_LENGTH, MODEL_ORDER, MODEL_TRUST_LEVEL, MODEL_VARIANTS, MatcherFormfactorDefinition, MemoryStorage, MissingToolsError, MultipleLlmExecutionTools, NAME, NonTaskSectionTypes, NotFoundError, NotYetImplementedError, ORDER_OF_PIPELINE_JSON, PLAYGROUND_APP_ID, PROMPTBOOK_ENGINE_VERSION, PROMPTBOOK_ERRORS, ParseError, PipelineExecutionError, PipelineLogicError, PipelineUrlError, PrefixStorage, PromptbookFetchError, REMOTE_SERVER_URLS, RESERVED_PARAMETER_NAMES, SET_IS_VERBOSE, SectionTypes, SheetsFormfactorDefinition, TaskTypes, TextFormatDefinition, TranslatorFormfactorDefinition, UNCERTAIN_USAGE, UNCERTAIN_ZERO_VALUE, UnexpectedError, WrappedError, ZERO_USAGE, ZERO_VALUE, _AnthropicClaudeMetadataRegistration, _AzureOpenAiMetadataRegistration, _BoilerplateScraperMetadataRegistration, _DeepseekMetadataRegistration, _DocumentScraperMetadataRegistration, _GoogleMetadataRegistration, _LegacyDocumentScraperMetadataRegistration, _MarkdownScraperMetadataRegistration, _MarkitdownScraperMetadataRegistration, _OpenAiAssistantMetadataRegistration, _OpenAiMetadataRegistration, _PdfScraperMetadataRegistration, _WebsiteScraperMetadataRegistration, addUsage, book, cacheLlmTools, collectionToJson, compilePipeline, countUsage, createCollectionFromJson, createCollectionFromPromise, createCollectionFromUrl, createLlmToolsFromConfiguration, createPipelineExecutor, createSubcollection, embeddingVectorToString, executionReportJsonToString, extractParameterNamesFromTask, filterModels, getPipelineInterface, identificationToPromptbookToken, isPassingExpectations, isPipelineImplementingInterface, isPipelineInterfacesEqual, isPipelinePrepared, isValidPipelineString, joinLlmExecutionTools, limitTotalUsage, makeKnowledgeSourceHandler, parsePipeline, pipelineJsonToString, prepareKnowledgePieces, preparePersona, preparePipeline, prepareTasks, prettifyPipelineString, promptbookFetch, promptbookTokenToIdentification, unpreparePipeline, usageToHuman, usageToWorktime, validatePipeline, validatePipelineString };
 //# sourceMappingURL=index.es.js.map