@promptbook/cli 0.61.0-24 → 0.61.0-25

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/README.md CHANGED
@@ -146,7 +146,7 @@ File `write-website-content.ptbk.md`:
146
146
  > - PROMPTBOOK VERSION 0.0.1
147
147
  > - INPUT  PARAM `{rawTitle}` Automatically suggested a site name or empty text
148
148
  > - INPUT  PARAM `{rawAssigment}` Automatically generated site entry from image recognition
149
- > - OUTPUT PARAM `{content}` Web content
149
+ > - OUTPUT PARAM `{websiteContent}` Web content
150
150
  > - OUTPUT PARAM `{keywords}` Keywords
151
151
  >
152
152
  > ## 👤 Specifying the assigment
@@ -305,7 +305,7 @@ File `write-website-content.ptbk.md`:
305
305
  > {contentBody}
306
306
  > ```
307
307
  >
308
- > `-> {content}`
308
+ > `-> {websiteContent}`
309
309
 
310
310
 
311
311
 
@@ -345,7 +345,7 @@ flowchart LR
345
345
  templateCombineTheBeginning--"{contentBeginning}"-->templateCombineTheContent
346
346
  templateWriteTheContent--"{contentBody}"-->templateCombineTheContent
347
347
 
348
- templateCombineTheContent--"{content}"-->output
348
+ templateCombineTheContent--"{websiteContent}"-->output
349
349
  output((Output)):::output
350
350
 
351
351
  classDef input color: grey;
package/esm/index.es.js CHANGED
@@ -150,7 +150,7 @@ new Function("\n try {\n if (typeof WorkerGlobalScope !== 'undefined'
150
150
  /**
151
151
  * The version of the Promptbook library
152
152
  */
153
- var PROMPTBOOK_VERSION = '0.61.0-23';
153
+ var PROMPTBOOK_VERSION = '0.61.0-24';
154
154
  // TODO: !!!! List here all the versions and annotate + put into script
155
155
 
156
156
  /**
@@ -287,6 +287,7 @@ var REPLACING_NONCE = 'u$k42k%!V2zo34w7Fu#@QUHYPW';
287
287
  * The names of the parameters that are reserved for special purposes
288
288
  */
289
289
  var RESERVED_PARAMETER_NAMES = deepFreeze([
290
+ 'content',
290
291
  'context',
291
292
  'knowledge',
292
293
  'samples',
@@ -299,6 +300,10 @@ var RESERVED_PARAMETER_NAMES = deepFreeze([
299
300
  * @@@
300
301
  */
301
302
  var RESERVED_PARAMETER_MISSING_VALUE = 'MISSING-' + REPLACING_NONCE;
303
+ /**
304
+ * @@@
305
+ */
306
+ var RESERVED_PARAMETER_RESTRICTED = 'RESTRICTED-' + REPLACING_NONCE;
302
307
  /*
303
308
  TODO: !!! Just testing false-negative detection of [🟡][🟢][🔵][⚪] leak
304
309
  */
@@ -742,7 +747,7 @@ function forEachAsync(array, options, callbackfunction) {
742
747
  });
743
748
  }
744
749
 
745
- var PipelineCollection = [{title:"Prepare Knowledge from Markdown",pipelineUrl:"https://promptbook.studio/promptbook/prepare-knowledge-from-markdown.ptbk.md",promptbookVersion:"0.61.0-23",parameters:[{name:"content",description:"Markdown document content",isInput:true,isOutput:false},{name:"knowledgePieces",description:"The knowledge JSON object",isInput:false,isOutput:true}],promptTemplates:[{blockType:"PROMPT_TEMPLATE",name:"knowledge",title:"Knowledge",modelRequirements:{modelVariant:"CHAT",modelName:"claude-3-opus-20240229"},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> {content}",dependentParameterNames:["content"],resultingParameterName:"knowledgePieces"}],knowledgeSources:[],knowledgePieces:[],personas:[],preparations:[{id:1,promptbookVersion:"0.61.0-23",modelUsage:{price:{value:0},input:{tokensCount:{value:0},charactersCount:{value:0},wordsCount:{value:0},sentencesCount:{value:0},linesCount:{value:0},paragraphsCount:{value:0},pagesCount:{value:0}},output:{tokensCount:{value:0},charactersCount:{value:0},wordsCount:{value:0},sentencesCount:{value:0},linesCount:{value:0},paragraphsCount:{value:0},pagesCount:{value:0}}}}],sourceFile:"./promptbook-collection/prepare-knowledge-from-markdown.ptbk.md"},{title:"Prepare Keywords",pipelineUrl:"https://promptbook.studio/promptbook/prepare-knowledge-keywords.ptbk.md",promptbookVersion:"0.61.0-23",parameters:[{name:"content",description:"The content",isInput:true,isOutput:false},{name:"keywords",description:"Keywords separated by comma",isInput:false,isOutput:true}],promptTemplates:[{blockType:"PROMPT_TEMPLATE",name:"knowledge",title:"Knowledge",modelRequirements:{modelVariant:"CHAT",modelName:"claude-3-opus-20240229"},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> {content}",dependentParameterNames:["content"],resultingParameterName:"keywords"}],knowledgeSources:[],knowledgePieces:[],personas:[],preparations:[{id:1,promptbookVersion:"0.61.0-23",modelUsage:{price:{value:0},input:{tokensCount:{value:0},charactersCount:{value:0},wordsCount:{value:0},sentencesCount:{value:0},linesCount:{value:0},paragraphsCount:{value:0},pagesCount:{value:0}},output:{tokensCount:{value:0},charactersCount:{value:0},wordsCount:{value:0},sentencesCount:{value:0},linesCount:{value:0},paragraphsCount:{value:0},pagesCount:{value:0}}}}],sourceFile:"./promptbook-collection/prepare-knowledge-keywords.ptbk.md"},{title:"Prepare Title",pipelineUrl:"https://promptbook.studio/promptbook/prepare-knowledge-title.ptbk.md",promptbookVersion:"0.61.0-23",parameters:[{name:"content",description:"The content",isInput:true,isOutput:false},{name:"title",description:"The title of the document",isInput:false,isOutput:true}],promptTemplates:[{blockType:"PROMPT_TEMPLATE",name:"knowledge",title:"Knowledge",modelRequirements:{modelVariant:"CHAT",modelName:"claude-3-opus-20240229"},content:"You are experienced content creator, write best title for the document.\n\n# Rules\n\n- Write just title, nothing else\n- Title should be concise and clear\n- Write maximum 5 words for the title\n\n# The document\n\n> {content}",expectations:{words:{min:1,max:8}},dependentParameterNames:["content"],resultingParameterName:"title"}],knowledgeSources:[],knowledgePieces:[],personas:[],preparations:[{id:1,promptbookVersion:"0.61.0-23",modelUsage:{price:{value:0},input:{tokensCount:{value:0},charactersCount:{value:0},wordsCount:{value:0},sentencesCount:{value:0},linesCount:{value:0},paragraphsCount:{value:0},pagesCount:{value:0}},output:{tokensCount:{value:0},charactersCount:{value:0},wordsCount:{value:0},sentencesCount:{value:0},linesCount:{value:0},paragraphsCount:{value:0},pagesCount:{value:0}}}}],sourceFile:"./promptbook-collection/prepare-knowledge-title.ptbk.md"},{title:"Prepare Keywords",pipelineUrl:"https://promptbook.studio/promptbook/prepare-persona.ptbk.md",promptbookVersion:"0.61.0-23",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}],promptTemplates:[{blockType:"PROMPT_TEMPLATE",name:"make-model-requirements",title:"Make modelRequirements",modelRequirements:{modelVariant:"CHAT",modelName:"gpt-4-turbo"},content:"You are experienced AI engineer, you need to create virtual assistant.\nWrite\n\n## Sample\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### Option `modelName`\n\nPick from the following models:\n\n- {availableModelNames}\n\n### Option `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### Option `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}",expectFormat:"JSON",dependentParameterNames:["availableModelNames","personaDescription"],resultingParameterName:"modelRequirements"}],knowledgeSources:[],knowledgePieces:[],personas:[],preparations:[{id:1,promptbookVersion:"0.61.0-23",modelUsage:{price:{value:0},input:{tokensCount:{value:0},charactersCount:{value:0},wordsCount:{value:0},sentencesCount:{value:0},linesCount:{value:0},paragraphsCount:{value:0},pagesCount:{value:0}},output:{tokensCount:{value:0},charactersCount:{value:0},wordsCount:{value:0},sentencesCount:{value:0},linesCount:{value:0},paragraphsCount:{value:0},pagesCount:{value:0}}}}],sourceFile:"./promptbook-collection/prepare-persona.ptbk.md"}];
750
+ var PipelineCollection = [{title:"Prepare Knowledge from Markdown",pipelineUrl:"https://promptbook.studio/promptbook/prepare-knowledge-from-markdown.ptbk.md",promptbookVersion:"0.61.0-24",parameters:[{name:"knowledgeContent",description:"Markdown document content",isInput:true,isOutput:false},{name:"knowledgePieces",description:"The knowledge JSON object",isInput:false,isOutput:true}],promptTemplates:[{blockType:"PROMPT_TEMPLATE",name:"knowledge",title:"Knowledge",modelRequirements:{modelVariant:"CHAT",modelName:"claude-3-opus-20240229"},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}",dependentParameterNames:["knowledgeContent"],resultingParameterName:"knowledgePieces"}],knowledgeSources:[],knowledgePieces:[],personas:[],preparations:[{id:1,promptbookVersion:"0.61.0-24",modelUsage:{price:{value:0},input:{tokensCount:{value:0},charactersCount:{value:0},wordsCount:{value:0},sentencesCount:{value:0},linesCount:{value:0},paragraphsCount:{value:0},pagesCount:{value:0}},output:{tokensCount:{value:0},charactersCount:{value:0},wordsCount:{value:0},sentencesCount:{value:0},linesCount:{value:0},paragraphsCount:{value:0},pagesCount:{value:0}}}}],sourceFile:"./promptbook-collection/prepare-knowledge-from-markdown.ptbk.md"},{title:"Prepare Keywords",pipelineUrl:"https://promptbook.studio/promptbook/prepare-knowledge-keywords.ptbk.md",promptbookVersion:"0.61.0-24",parameters:[{name:"knowledgePieceContent",description:"The content",isInput:true,isOutput:false},{name:"keywords",description:"Keywords separated by comma",isInput:false,isOutput:true}],promptTemplates:[{blockType:"PROMPT_TEMPLATE",name:"knowledge",title:"Knowledge",modelRequirements:{modelVariant:"CHAT",modelName:"claude-3-opus-20240229"},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}",dependentParameterNames:["knowledgePieceContent"],resultingParameterName:"keywords"}],knowledgeSources:[],knowledgePieces:[],personas:[],preparations:[{id:1,promptbookVersion:"0.61.0-24",modelUsage:{price:{value:0},input:{tokensCount:{value:0},charactersCount:{value:0},wordsCount:{value:0},sentencesCount:{value:0},linesCount:{value:0},paragraphsCount:{value:0},pagesCount:{value:0}},output:{tokensCount:{value:0},charactersCount:{value:0},wordsCount:{value:0},sentencesCount:{value:0},linesCount:{value:0},paragraphsCount:{value:0},pagesCount:{value:0}}}}],sourceFile:"./promptbook-collection/prepare-knowledge-keywords.ptbk.md"},{title:"Prepare Title",pipelineUrl:"https://promptbook.studio/promptbook/prepare-knowledge-title.ptbk.md",promptbookVersion:"0.61.0-24",parameters:[{name:"knowledgePieceContent",description:"The content",isInput:true,isOutput:false},{name:"title",description:"The title of the document",isInput:false,isOutput:true}],promptTemplates:[{blockType:"PROMPT_TEMPLATE",name:"knowledge",title:"Knowledge",modelRequirements:{modelVariant:"CHAT",modelName:"claude-3-opus-20240229"},content:"You are experienced content creator, write best title for the document.\n\n# Rules\n\n- Write just title, nothing else\n- Title should be concise and clear\n- Write maximum 5 words for the title\n\n# The document\n\n> {knowledgePieceContent}",expectations:{words:{min:1,max:8}},dependentParameterNames:["knowledgePieceContent"],resultingParameterName:"title"}],knowledgeSources:[],knowledgePieces:[],personas:[],preparations:[{id:1,promptbookVersion:"0.61.0-24",modelUsage:{price:{value:0},input:{tokensCount:{value:0},charactersCount:{value:0},wordsCount:{value:0},sentencesCount:{value:0},linesCount:{value:0},paragraphsCount:{value:0},pagesCount:{value:0}},output:{tokensCount:{value:0},charactersCount:{value:0},wordsCount:{value:0},sentencesCount:{value:0},linesCount:{value:0},paragraphsCount:{value:0},pagesCount:{value:0}}}}],sourceFile:"./promptbook-collection/prepare-knowledge-title.ptbk.md"},{title:"Prepare Keywords",pipelineUrl:"https://promptbook.studio/promptbook/prepare-persona.ptbk.md",promptbookVersion:"0.61.0-24",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}],promptTemplates:[{blockType:"PROMPT_TEMPLATE",name:"make-model-requirements",title:"Make modelRequirements",modelRequirements:{modelVariant:"CHAT",modelName:"gpt-4-turbo"},content:"You are experienced AI engineer, you need to create virtual assistant.\nWrite\n\n## Sample\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### Option `modelName`\n\nPick from the following models:\n\n- {availableModelNames}\n\n### Option `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### Option `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}",expectFormat:"JSON",dependentParameterNames:["availableModelNames","personaDescription"],resultingParameterName:"modelRequirements"}],knowledgeSources:[],knowledgePieces:[],personas:[],preparations:[{id:1,promptbookVersion:"0.61.0-24",modelUsage:{price:{value:0},input:{tokensCount:{value:0},charactersCount:{value:0},wordsCount:{value:0},sentencesCount:{value:0},linesCount:{value:0},paragraphsCount:{value:0},pagesCount:{value:0}},output:{tokensCount:{value:0},charactersCount:{value:0},wordsCount:{value:0},sentencesCount:{value:0},linesCount:{value:0},paragraphsCount:{value:0},pagesCount:{value:0}}}}],sourceFile:"./promptbook-collection/prepare-persona.ptbk.md"}];
746
751
 
747
752
  /**
748
753
  * This error indicates that the promptbook in a markdown format cannot be parsed into a valid promptbook object
@@ -1176,6 +1181,36 @@ var ReferenceError$1 = /** @class */ (function (_super) {
1176
1181
  return ReferenceError;
1177
1182
  }(Error));
1178
1183
 
1184
+ /**
1185
+ * Parses the template and returns the list of all parameter names
1186
+ *
1187
+ * @param template the template with parameters in {curly} braces
1188
+ * @returns the list of parameter names
1189
+ */
1190
+ function extractParameters(template) {
1191
+ var e_1, _a;
1192
+ var matches = template.matchAll(/{\w+}/g);
1193
+ var parameterNames = new Set();
1194
+ try {
1195
+ for (var matches_1 = __values(matches), matches_1_1 = matches_1.next(); !matches_1_1.done; matches_1_1 = matches_1.next()) {
1196
+ var match = matches_1_1.value;
1197
+ var parameterName = match[0].slice(1, -1);
1198
+ parameterNames.add(parameterName);
1199
+ }
1200
+ }
1201
+ catch (e_1_1) { e_1 = { error: e_1_1 }; }
1202
+ finally {
1203
+ try {
1204
+ if (matches_1_1 && !matches_1_1.done && (_a = matches_1.return)) _a.call(matches_1);
1205
+ }
1206
+ finally { if (e_1) throw e_1.error; }
1207
+ }
1208
+ return parameterNames;
1209
+ }
1210
+ /**
1211
+ * TODO: !!!!! Rename to extractParameterNames
1212
+ */
1213
+
1179
1214
  /**
1180
1215
  * Unprepare just strips the preparation data of the pipeline
1181
1216
  */
@@ -1183,7 +1218,14 @@ function unpreparePipeline(pipeline) {
1183
1218
  var personas = pipeline.personas, knowledgeSources = pipeline.knowledgeSources, promptTemplates = pipeline.promptTemplates;
1184
1219
  personas = personas.map(function (persona) { return (__assign(__assign({}, persona), { modelRequirements: undefined, preparationIds: undefined })); });
1185
1220
  knowledgeSources = knowledgeSources.map(function (knowledgeSource) { return (__assign(__assign({}, knowledgeSource), { preparationIds: undefined })); });
1186
- promptTemplates = promptTemplates.map(function (promptTemplate) { return (__assign(__assign({}, promptTemplate), { preparedContent: undefined })); });
1221
+ promptTemplates = promptTemplates.map(function (promptTemplate) {
1222
+ var dependentParameterNames = promptTemplate.dependentParameterNames;
1223
+ var parameterNames = extractParameters(promptTemplate.preparedContent || '');
1224
+ dependentParameterNames = dependentParameterNames.filter(function (dependentParameterName) { return !parameterNames.has(dependentParameterName); });
1225
+ var promptTemplateUnprepared = __assign(__assign({}, promptTemplate), { dependentParameterNames: dependentParameterNames });
1226
+ delete promptTemplateUnprepared.preparedContent;
1227
+ return promptTemplateUnprepared;
1228
+ });
1187
1229
  return __assign(__assign({}, pipeline), { promptTemplates: promptTemplates, knowledgeSources: knowledgeSources, knowledgePieces: [], personas: personas, preparations: [] });
1188
1230
  }
1189
1231
  /**
@@ -1680,33 +1722,6 @@ function assertsExecutionSuccessful(executionResult) {
1680
1722
  * TODO: [🧠] Can this return type be better typed than void
1681
1723
  */
1682
1724
 
1683
- /**
1684
- * Parses the template and returns the list of all parameter names
1685
- *
1686
- * @param template the template with parameters in {curly} braces
1687
- * @returns the list of parameter names
1688
- */
1689
- function extractParameters(template) {
1690
- var e_1, _a;
1691
- var matches = template.matchAll(/{\w+}/g);
1692
- var parameterNames = new Set();
1693
- try {
1694
- for (var matches_1 = __values(matches), matches_1_1 = matches_1.next(); !matches_1_1.done; matches_1_1 = matches_1.next()) {
1695
- var match = matches_1_1.value;
1696
- var parameterName = match[0].slice(1, -1);
1697
- parameterNames.add(parameterName);
1698
- }
1699
- }
1700
- catch (e_1_1) { e_1 = { error: e_1_1 }; }
1701
- finally {
1702
- try {
1703
- if (matches_1_1 && !matches_1_1.done && (_a = matches_1.return)) _a.call(matches_1);
1704
- }
1705
- finally { if (e_1) throw e_1.error; }
1706
- }
1707
- return parameterNames;
1708
- }
1709
-
1710
1725
  /**
1711
1726
  * Parses the given script and returns the list of all used variables that are not defined in the script
1712
1727
  *
@@ -1764,10 +1779,10 @@ function extractVariables(script) {
1764
1779
  */
1765
1780
  function extractParametersFromPromptTemplate(promptTemplate) {
1766
1781
  var e_1, _a, e_2, _b, e_3, _c;
1767
- var title = promptTemplate.title, description = promptTemplate.description, blockType = promptTemplate.blockType, content = promptTemplate.content, jokerParameterNames = promptTemplate.jokerParameterNames;
1782
+ var title = promptTemplate.title, description = promptTemplate.description, blockType = promptTemplate.blockType, content = promptTemplate.content, preparedContent = promptTemplate.preparedContent, jokerParameterNames = promptTemplate.jokerParameterNames;
1768
1783
  var parameterNames = new Set();
1769
1784
  try {
1770
- for (var _d = __values(__spreadArray(__spreadArray(__spreadArray([], __read(extractParameters(title)), false), __read(extractParameters(description || '')), false), __read(extractParameters(content)), false)), _e = _d.next(); !_e.done; _e = _d.next()) {
1785
+ for (var _d = __values(__spreadArray(__spreadArray(__spreadArray(__spreadArray([], __read(extractParameters(title)), false), __read(extractParameters(description || '')), false), __read(extractParameters(content)), false), __read(extractParameters(preparedContent || '')), false)), _e = _d.next(); !_e.done; _e = _d.next()) {
1771
1786
  var parameterName = _e.value;
1772
1787
  parameterNames.add(parameterName);
1773
1788
  }
@@ -1807,6 +1822,8 @@ function extractParametersFromPromptTemplate(promptTemplate) {
1807
1822
  }
1808
1823
  finally { if (e_3) throw e_3.error; }
1809
1824
  }
1825
+ parameterNames.delete('content');
1826
+ // <- Note {websiteContent} is used in `preparedContent`
1810
1827
  return parameterNames;
1811
1828
  }
1812
1829
  /**
@@ -1976,7 +1993,14 @@ var MultipleLlmExecutionTools = /** @class */ (function () {
1976
1993
  throw errors[0];
1977
1994
  }
1978
1995
  else if (errors.length > 1) {
1979
- throw new PipelineExecutionError(spaceTrim(function (block) { return "\n All execution tools failed:\n\n ".concat(block(errors.map(function (error) { return "- ".concat(error.name || 'Error', ": ").concat(error.message); }).join('\n')), "\n\n "); }));
1996
+ throw new PipelineExecutionError(
1997
+ // TODO: Tell which execution tools failed like
1998
+ // 1) OpenAI throw PipelineExecutionError: Parameter {knowledge} is not defined
1999
+ // 2) AnthropicClaude throw PipelineExecutionError: Parameter {knowledge} is not defined
2000
+ // 3) ...
2001
+ spaceTrim(function (block) { return "\n All execution tools failed:\n\n ".concat(block(errors
2002
+ .map(function (error, i) { return "".concat(i + 1, ") **").concat(error.name || 'Error', ":** ").concat(error.message); })
2003
+ .join('\n')), "\n\n "); }));
1980
2004
  }
1981
2005
  else {
1982
2006
  throw new PipelineExecutionError(spaceTrim(function (block) { return "\n You have not provided any `LlmExecutionTools` that support model variant \"".concat(prompt.modelRequirements.modelVariant, "\n\n Available `LlmExecutionTools`:\n ").concat(block(_this.llmExecutionTools
@@ -2181,6 +2205,10 @@ function replaceParameters(template, parameters) {
2181
2205
  if (parameterValue === RESERVED_PARAMETER_MISSING_VALUE) {
2182
2206
  throw new UnexpectedError("Parameter {".concat(parameterName, "} has missing value"));
2183
2207
  }
2208
+ else if (parameterValue === RESERVED_PARAMETER_RESTRICTED) {
2209
+ // TODO: [🍵]
2210
+ throw new UnexpectedError("Parameter {".concat(parameterName, "} is restricted to use"));
2211
+ }
2184
2212
  }
2185
2213
  }
2186
2214
  catch (e_1_1) { e_1 = { error: e_1_1 }; }
@@ -2443,7 +2471,7 @@ function createPipelineExecutor(options) {
2443
2471
  console.warn(spaceTrim$1("\n Pipeline ".concat(rawPipeline.pipelineUrl || rawPipeline.sourceFile || rawPipeline.title, " is not prepared\n\n ").concat(rawPipeline.sourceFile, "\n\n It will be prepared ad-hoc before the first execution\n But it is recommended to prepare the pipeline during collection preparation\n\n @see more at https://ptbk.io/prepare-pipeline\n ")));
2444
2472
  }
2445
2473
  var pipelineExecutor = function (inputParameters, onProgress) { return __awaiter(_this, void 0, void 0, function () {
2446
- // TODO: !!!!! Extract to separate functions and files - ALL FUNCTIONS BELOW
2474
+ // TODO: !!! Extract to separate functions and files - ALL FUNCTIONS BELOW
2447
2475
  function getContextForTemplate(// <- TODO: [🧠][🥜]
2448
2476
  template) {
2449
2477
  return __awaiter(this, void 0, void 0, function () {
@@ -2494,6 +2522,7 @@ function createPipelineExecutor(options) {
2494
2522
  currentDate = new Date().toISOString();
2495
2523
  modelName = RESERVED_PARAMETER_MISSING_VALUE;
2496
2524
  reservedParameters = {
2525
+ content: RESERVED_PARAMETER_RESTRICTED,
2497
2526
  context: context,
2498
2527
  knowledge: knowledge,
2499
2528
  samples: samples,
@@ -3204,6 +3233,8 @@ function createPipelineExecutor(options) {
3204
3233
  return pipelineExecutor;
3205
3234
  }
3206
3235
  /**
3236
+ * TODO: !!!! return `preparedPipeline` from execution
3237
+ * TODO: !!!! `isNotPreparedWarningSupressed`
3207
3238
  * TODO: Use isVerbose here (not only pass to `preparePipeline`)
3208
3239
  * TODO: [🪂] Use maxParallelCount here (not only pass to `preparePipeline`)
3209
3240
  * TODO: [♈] Probbably move expectations from templates to parameters
@@ -3218,7 +3249,7 @@ function createPipelineExecutor(options) {
3218
3249
  /**
3219
3250
  * @@@
3220
3251
  */
3221
- function prepareKnowledgeFromMarkdown(content /* <- TODO: [🖖] (?maybe not) Always the file */, options) {
3252
+ function prepareKnowledgeFromMarkdown(knowledgeContent /* <- TODO: [🖖] (?maybe not) Always the file */, options) {
3222
3253
  return __awaiter(this, void 0, void 0, function () {
3223
3254
  var llmTools, _a, maxParallelCount, _b, isVerbose, collection, prepareKnowledgeFromMarkdownExecutor, _c, prepareTitleExecutor, _d, prepareKeywordsExecutor, _e, result, outputParameters, knowledgePiecesRaw, knowledgeTextPieces, knowledge;
3224
3255
  var _f, _g, _h;
@@ -3256,7 +3287,7 @@ function prepareKnowledgeFromMarkdown(content /* <- TODO: [🖖] (?maybe not) Al
3256
3287
  llm: llmTools,
3257
3288
  },
3258
3289
  _h)]);
3259
- return [4 /*yield*/, prepareKnowledgeFromMarkdownExecutor({ content: content })];
3290
+ return [4 /*yield*/, prepareKnowledgeFromMarkdownExecutor({ knowledgeContent: knowledgeContent })];
3260
3291
  case 4:
3261
3292
  result = _j.sent();
3262
3293
  assertsExecutionSuccessful(result);
@@ -3269,25 +3300,25 @@ function prepareKnowledgeFromMarkdown(content /* <- TODO: [🖖] (?maybe not) Al
3269
3300
  return [4 /*yield*/, Promise.all(
3270
3301
  // TODO: [🪂] !! Do not send all at once but in chunks
3271
3302
  knowledgeTextPieces.map(function (knowledgeTextPiece, i) { return __awaiter(_this, void 0, void 0, function () {
3272
- var name, title, content, keywords, index, titleResult, _a, titleRaw, keywordsResult, _b, keywordsRaw, embeddingResult, error_1;
3303
+ var name, title, knowledgePieceContent, keywords, index, titleResult, _a, titleRaw, keywordsResult, _b, keywordsRaw, embeddingResult, error_1;
3273
3304
  return __generator(this, function (_c) {
3274
3305
  switch (_c.label) {
3275
3306
  case 0:
3276
3307
  name = "piece-".concat(i);
3277
3308
  title = spaceTrim(knowledgeTextPiece.substring(0, 100));
3278
- content = spaceTrim(knowledgeTextPiece);
3309
+ knowledgePieceContent = spaceTrim(knowledgeTextPiece);
3279
3310
  keywords = [];
3280
3311
  index = [];
3281
3312
  _c.label = 1;
3282
3313
  case 1:
3283
3314
  _c.trys.push([1, 7, , 8]);
3284
- return [4 /*yield*/, prepareTitleExecutor({ content: content })];
3315
+ return [4 /*yield*/, prepareTitleExecutor({ knowledgePieceContent: knowledgePieceContent })];
3285
3316
  case 2:
3286
3317
  titleResult = _c.sent();
3287
3318
  _a = titleResult.outputParameters.title, titleRaw = _a === void 0 ? 'Untitled' : _a;
3288
3319
  title = spaceTrim(titleRaw) /* <- TODO: Maybe do in pipeline */;
3289
3320
  name = titleToName(title);
3290
- return [4 /*yield*/, prepareKeywordsExecutor({ content: content })];
3321
+ return [4 /*yield*/, prepareKeywordsExecutor({ knowledgePieceContent: knowledgePieceContent })];
3291
3322
  case 3:
3292
3323
  keywordsResult = _c.sent();
3293
3324
  _b = keywordsResult.outputParameters.keywords, keywordsRaw = _b === void 0 ? '' : _b;
@@ -3305,7 +3336,7 @@ function prepareKnowledgeFromMarkdown(content /* <- TODO: [🖖] (?maybe not) Al
3305
3336
  case 4: return [4 /*yield*/, llmTools.callEmbeddingModel({
3306
3337
  title: "Embedding for ".concat(title) /* <- Note: No impact on embedding result itself, just for logging */,
3307
3338
  parameters: {},
3308
- content: content,
3339
+ content: knowledgePieceContent,
3309
3340
  modelRequirements: {
3310
3341
  modelVariant: 'EMBEDDING',
3311
3342
  },
@@ -3326,7 +3357,7 @@ function prepareKnowledgeFromMarkdown(content /* <- TODO: [🖖] (?maybe not) Al
3326
3357
  case 8: return [2 /*return*/, {
3327
3358
  name: name,
3328
3359
  title: title,
3329
- content: content,
3360
+ content: knowledgePieceContent,
3330
3361
  keywords: keywords,
3331
3362
  index: index,
3332
3363
  // <- TODO: [☀] sources,
@@ -3492,14 +3523,18 @@ function prepareTemplates(pipeline, options) {
3492
3523
  TODO_USE(parameters);
3493
3524
  promptTemplatesPrepared = new Array(promptTemplates.length);
3494
3525
  return [4 /*yield*/, forEachAsync(promptTemplates, { maxParallelCount: maxParallelCount /* <- TODO: [🪂] When there are subtasks, this maximul limit can be broken */ }, function (template, index) { return __awaiter(_this, void 0, void 0, function () {
3495
- var preparedContent, preparedTemplate;
3526
+ var dependentParameterNames, preparedContent, preparedTemplate;
3496
3527
  return __generator(this, function (_a) {
3528
+ dependentParameterNames = template.dependentParameterNames;
3497
3529
  preparedContent = undefined;
3498
- if (knowledgePiecesCount > 0) {
3530
+ if (knowledgePiecesCount > 0 && !dependentParameterNames.includes('knowledge')) {
3499
3531
  preparedContent = spaceTrim$1("\n {content}\n\n ## Knowledge\n\n {knowledge}\n ");
3500
3532
  // <- TODO: [🧠][🧻] Cutomize shape/language/formatting of the addition to the prompt
3533
+ dependentParameterNames = __spreadArray(__spreadArray([], __read(dependentParameterNames), false), [
3534
+ 'knowledge',
3535
+ ], false);
3501
3536
  }
3502
- preparedTemplate = __assign(__assign({}, template), { preparedContent: preparedContent });
3537
+ preparedTemplate = __assign(__assign({}, template), { dependentParameterNames: dependentParameterNames, preparedContent: preparedContent });
3503
3538
  promptTemplatesPrepared[index] = preparedTemplate;
3504
3539
  return [2 /*return*/];
3505
3540
  });
@@ -4315,7 +4350,7 @@ var parameterCommandParser = {
4315
4350
  /**
4316
4351
  * Example usages of the PARAMETER command
4317
4352
  */
4318
- examples: ['PARAMETER {title} Title of the book', 'OUTPUT PARAMETER {content} Content of the book'],
4353
+ examples: ['PARAMETER {title} Title of the book', 'OUTPUT PARAMETER {websiteContent} Content of the book'],
4319
4354
  /**
4320
4355
  * Parses the PARAMETER command
4321
4356
  */