@promptbook/wizard 0.95.0

package/esm/typings/src/scrapers/_common/register/ScraperAndConverterMetadata.d.ts

@@ -0,0 +1,46 @@
+import type { Executables } from '../../../execution/Executables';
+import type { string_mime_type } from '../../../types/typeAliases';
+import type { string_promptbook_documentation_url } from '../../../types/typeAliases';
+import type { string_title } from '../../../types/typeAliases';
+import type { Registered } from '../../../utils/$Register';
+/**
+ * Metadata interface for scrapers and converters in the system.
+ * Contains information about the capabilities and requirements of a scraper or converter.
+ *
+ * x) `Scraper`
+ * x) `Converter`
+ * x) `ScraperConstructor`
+ * x) `Registered`
+ * x) `ExecutionTools`
+ * x) `ScraperAndConverterMetadata`
+ * x) `PrepareAndScrapeOptions`
+ * x) `ScraperConfiguration`
+ * x) `ScraperOptions`
+ */
+export type ScraperAndConverterMetadata = Registered & {
+    /**
+     * Human-readable title of the scraper or converter.
+     * Used for display purposes in logs and interfaces.
+     */
+    readonly title: string_title;
+    /**
+     * Mime types that this scraper can handle
+     */
+    readonly mimeTypes: ReadonlyArray<string_mime_type>;
+    /**
+     * Flag indicating whether this scraper or converter can run in a browser environment.
+     * Some scrapers require Node.js capabilities and cannot run client-side.
+     *
+     * Note: [🌏] Only `MarkdownScraper` makes sense to be available in the browser, for scraping non-markdown sources in the browser use a remote server
+     */
+    readonly isAvailableInBrowser: boolean;
+    /**
+     * List of executables required by this scraper or converter to function properly.
+     * For example, PDF scrapers may require 'pandoc' to be installed on the system.
+     */
+    readonly requiredExecutables: ReadonlyArray<Capitalize<keyof Executables extends `${infer N}Path` ? N : never>>;
+    /**
+     * Link to documentation
+     */
+    readonly documentationUrl: string_promptbook_documentation_url;
+};

package/esm/typings/src/scrapers/_common/register/ScraperConstructor.d.ts

@@ -0,0 +1,13 @@
+import type { ExecutionTools } from '../../../execution/ExecutionTools';
+import type { PrepareAndScrapeOptions } from '../../../prepare/PrepareAndScrapeOptions';
+import type { Registered } from '../../../utils/$Register';
+import type { Scraper } from '../Scraper';
+import type { ScraperAndConverterMetadata } from './ScraperAndConverterMetadata';
+/**
+ * Type definition for a constructor function that creates a Scraper instance.
+ * Used for registering scrapers in the system to handle different content types.
+ */
+export type ScraperConstructor = Registered & ScraperAndConverterMetadata & ((tools: Pick<ExecutionTools, 'llm'>, options: PrepareAndScrapeOptions) => Scraper);
+/**
+ * TODO: [🎶] Naming "constructor" vs "creator" vs "factory"
+ */

package/esm/typings/src/scrapers/_common/utils/getScraperIntermediateSource.d.ts

@@ -0,0 +1,35 @@
+import type { PrepareAndScrapeOptions } from '../../../prepare/PrepareAndScrapeOptions';
+import type { string_file_extension } from '../../../types/typeAliases';
+import type { ScraperSourceHandler } from '../Scraper';
+import type { ScraperIntermediateSource } from '../ScraperIntermediateSource';
+/**
+ * Type representing source information for generating intermediate storage.
+ * Contains essential file and URL information needed to create a deterministic storage path.
+ *
+ * @private internal utility of `getScraperIntermediateSource`
+ */
+type GetScraperIntermediateSourceSource = Pick<ScraperSourceHandler, 'filename' | 'url'>;
+/**
+ * Options for intermediate source generation and management.
+ * Configures how intermediate files are named, stored, and cleaned up.
+ *
+ * @private internal utility of `getScraperIntermediateSource`
+ */
+type GetScraperIntermediateSourceHandler = Required<Pick<PrepareAndScrapeOptions, 'rootDirname' | 'cacheDirname' | 'intermediateFilesStrategy' | 'isVerbose'>> & {
+    readonly extension: string_file_extension;
+};
+/**
+ * Retrieves an intermediate source for a scraper based on the knowledge source.
+ * Manages the caching and retrieval of intermediate scraper results for optimized performance.
+ *
+ * @private as internal utility for scrapers
+ */
+export declare function getScraperIntermediateSource(source: GetScraperIntermediateSourceSource, options: GetScraperIntermediateSourceHandler): Promise<ScraperIntermediateSource>;
+export {};
+/**
+ * Note: Not using `FileCacheStorage` for two reasons:
+ * 1) Need to store more than serialized JSONs
+ * 2) Need to switch between a `rootDirname` and `cacheDirname` <- TODO: [😡]
+ * TODO: [🐱‍🐉][🧠] Make some smart crop
+ * Note: [🟢] Code in this file should never be never released in packages that could be imported into browser environment
+ */

package/esm/typings/src/scrapers/_common/utils/getScraperIntermediateSource.test.d.ts

@@ -0,0 +1,4 @@
+export {};
+/**
+ * TODO: [🐱‍🐉][🧠] Make some smart crop
+ */

package/esm/typings/src/scrapers/_common/utils/makeKnowledgeSourceHandler.d.ts

@@ -0,0 +1,13 @@
+import type { SetOptional } from 'type-fest';
+import type { ExecutionTools } from '../../../execution/ExecutionTools';
+import type { KnowledgeSourceJson } from '../../../pipeline/PipelineJson/KnowledgeSourceJson';
+import type { PrepareAndScrapeOptions } from '../../../prepare/PrepareAndScrapeOptions';
+import type { ScraperSourceHandler } from '../Scraper';
+/**
+ * Factory function that creates a handler for processing knowledge sources.
+ * Provides standardized processing of different types of knowledge sources
+ * across various scraper implementations.
+ *
+ * @public exported from `@promptbook/core`
+ */
+export declare function makeKnowledgeSourceHandler(knowledgeSource: SetOptional<KnowledgeSourceJson, 'name'>, tools: Pick<ExecutionTools, 'fs' | 'fetch'>, options?: Pick<PrepareAndScrapeOptions, 'rootDirname' | 'isVerbose'>): Promise<ScraperSourceHandler>;

package/esm/typings/src/scrapers/_common/utils/promptbookFetch.d.ts

@@ -0,0 +1,10 @@
+import type { PromptbookFetch } from '../../../execution/PromptbookFetch';
+/**
+ * The built-in `fetch' function with a lightweight error handling wrapper as default fetch function used in Promptbook scrapers
+ *
+ * @public exported from `@promptbook/core`
+ */
+export declare const promptbookFetch: PromptbookFetch;
+/**
+ * TODO: [🧠] Maybe rename because it is not used only for scrapers but also in `$getCompiledBook`
+ */

package/esm/typings/src/scrapers/document/DocumentScraper.d.ts

@@ -0,0 +1,43 @@
+import type { KnowledgePiecePreparedJson } from '../../pipeline/PipelineJson/KnowledgePieceJson';
+import type { ExecutionTools } from '../../execution/ExecutionTools';
+import type { PrepareAndScrapeOptions } from '../../prepare/PrepareAndScrapeOptions';
+import type { Converter } from '../_common/Converter';
+import type { ScraperAndConverterMetadata } from '../_common/register/ScraperAndConverterMetadata';
+import type { Scraper } from '../_common/Scraper';
+import type { ScraperSourceHandler } from '../_common/Scraper';
+import type { ScraperIntermediateSource } from '../_common/ScraperIntermediateSource';
+/**
+ * Scraper of .docx and .odt files
+ *
+ * @see `documentationUrl` for more details
+ * @public exported from `@promptbook/documents`
+ */
+export declare class DocumentScraper implements Converter, Scraper {
+    private readonly tools;
+    private readonly options;
+    /**
+     * Metadata of the scraper which includes title, mime types, etc.
+     */
+    get metadata(): ScraperAndConverterMetadata;
+    /**
+     * Markdown scraper is used internally
+     */
+    private readonly markdownScraper;
+    constructor(tools: Pick<ExecutionTools, 'fs' | 'llm' | 'executables'>, options: PrepareAndScrapeOptions);
+    /**
+     * Convert the `.docx` or `.odt`  to `.md` file and returns intermediate source
+     *
+     * Note: `$` is used to indicate that this function is not a pure function - it leaves files on the disk and you are responsible for cleaning them by calling `destroy` method of returned object
+     */
+    $convert(source: ScraperSourceHandler): Promise<ScraperIntermediateSource>;
+    /**
+     * Scrapes the docx file and returns the knowledge pieces or `null` if it can't scrape it
+     */
+    scrape(source: ScraperSourceHandler): Promise<ReadonlyArray<Omit<KnowledgePiecePreparedJson, 'sources' | 'preparationIds'>> | null>;
+}
+/**
+ * TODO: [👣] Converted documents can act as cached items - there is no need to run conversion each time
+ * TODO: [🪂] Do it in parallel 11:11
+ * Note: No need to aggregate usage here, it is done by intercepting the llmTools
+ * Note: [🟢] Code in this file should never be never released in packages that could be imported into browser environment
+ */

package/esm/typings/src/scrapers/document/DocumentScraper.test.d.ts

@@ -0,0 +1,4 @@
+export {};
+/**
+ * TODO: [📓] Maybe test all file in examples (not just 10-simple.docx)
+ */

package/esm/typings/src/scrapers/document/createDocumentScraper.d.ts

@@ -0,0 +1,20 @@
+import type { ExecutionTools } from '../../execution/ExecutionTools';
+import type { PrepareAndScrapeOptions } from '../../prepare/PrepareAndScrapeOptions';
+import { DocumentScraper } from './DocumentScraper';
+/**
+ * Creates a scraper for document content.
+ *
+ * @public exported from `@promptbook/documents`
+ */
+export declare const createDocumentScraper: ((tools: Pick<ExecutionTools, 'llm'>, options: PrepareAndScrapeOptions) => DocumentScraper) & import("type-fest/source/readonly-deep").ReadonlyObjectDeep<{
+    title: string;
+    packageName: string;
+    className: string;
+    mimeTypes: string[];
+    documentationUrl: "https://github.com/webgptorg/promptbook/discussions/@@";
+    isAvailableInBrowser: false;
+    requiredExecutables: "Pandoc"[];
+}>;
+/**
+ * TODO: [🎶] Naming "constructor" vs "creator" vs "factory"
+ */

package/esm/typings/src/scrapers/document/playground/document-scraper-playground.d.ts

@@ -0,0 +1,5 @@
+#!/usr/bin/env ts-node
+export {};
+/**
+ * Note: [⚫] Code in this file should never be published in any package
+ */

package/esm/typings/src/scrapers/document/register-constructor.d.ts

@@ -0,0 +1,15 @@
+import type { Registration } from '../../utils/$Register';
+/**
+ * Registration of known scraper
+ *
+ * Warning: This is not useful for the end user, it is just a side effect of the mechanism that handles all available known scrapers
+ *
+ * @public exported from `@promptbook/documents`
+ * @public exported from `@promptbook/wizard`
+ * @public exported from `@promptbook/cli`
+ */
+export declare const _DocumentScraperRegistration: Registration;
+/**
+ * TODO: [🎶] Naming "constructor" vs "creator" vs "factory"
+ * Note: [💞] Ignore a discrepancy between file name and entity name
+ */

package/esm/typings/src/scrapers/document/register-metadata.d.ts

@@ -0,0 +1,28 @@
+import type { Registration } from '../../utils/$Register';
+/**
+ * Metadata of the scraper
+ *
+ * @private within the scraper directory
+ */
+export declare const documentScraperMetadata: import("type-fest/source/readonly-deep").ReadonlyObjectDeep<{
+    title: string;
+    packageName: string;
+    className: string;
+    mimeTypes: string[];
+    documentationUrl: "https://github.com/webgptorg/promptbook/discussions/@@";
+    isAvailableInBrowser: false;
+    requiredExecutables: "Pandoc"[];
+}>;
+/**
+ * Registration of known scraper metadata
+ *
+ * Warning: This is not useful for the end user, it is just a side effect of the mechanism that handles all available known scrapers
+ *
+ * @public exported from `@promptbook/core`
+ * @public exported from `@promptbook/wizard`
+ * @public exported from `@promptbook/cli`
+ */
+export declare const _DocumentScraperMetadataRegistration: Registration;
+/**
+ * Note: [💞] Ignore a discrepancy between file name and entity name
+ */

package/esm/typings/src/scrapers/document-legacy/LegacyDocumentScraper.d.ts

@@ -0,0 +1,43 @@
+import type { KnowledgePiecePreparedJson } from '../../pipeline/PipelineJson/KnowledgePieceJson';
+import type { ExecutionTools } from '../../execution/ExecutionTools';
+import type { PrepareAndScrapeOptions } from '../../prepare/PrepareAndScrapeOptions';
+import type { Converter } from '../_common/Converter';
+import type { ScraperAndConverterMetadata } from '../_common/register/ScraperAndConverterMetadata';
+import type { Scraper } from '../_common/Scraper';
+import type { ScraperSourceHandler } from '../_common/Scraper';
+import type { ScraperIntermediateSource } from '../_common/ScraperIntermediateSource';
+/**
+ * Scraper for old document files (like .doc and .rtf)
+ *
+ * @see `documentationUrl` for more details
+ * @public exported from `@promptbook/legacy-documents`
+ */
+export declare class LegacyDocumentScraper implements Converter, Scraper {
+    private readonly tools;
+    private readonly options;
+    /**
+     * Metadata of the scraper which includes title, mime types, etc.
+     */
+    get metadata(): ScraperAndConverterMetadata;
+    /**
+     * Document scraper is used internally
+     */
+    private readonly documentScraper;
+    constructor(tools: Pick<ExecutionTools, 'fs' | 'llm' | 'executables'>, options: PrepareAndScrapeOptions);
+    /**
+     * Convert the `.doc` or `.rtf`  to `.doc` file and returns intermediate source
+     *
+     * Note: `$` is used to indicate that this function is not a pure function - it leaves files on the disk and you are responsible for cleaning them by calling `destroy` method of returned object
+     */
+    $convert(source: ScraperSourceHandler): Promise<ScraperIntermediateSource>;
+    /**
+     * Scrapes the `.doc` or `.rtf` file and returns the knowledge pieces or `null` if it can't scrape it
+     */
+    scrape(source: ScraperSourceHandler): Promise<ReadonlyArray<Omit<KnowledgePiecePreparedJson, 'sources' | 'preparationIds'>> | null>;
+}
+/**
+ * TODO: [👣] Converted documents can act as cached items - there is no need to run conversion each time
+ * TODO: [🪂] Do it in parallel 11:11
+ * Note: No need to aggregate usage here, it is done by intercepting the llmTools
+ * Note: [🟢] Code in this file should never be never released in packages that could be imported into browser environment
+ */

package/esm/typings/src/scrapers/document-legacy/LegacyDocumentScraper.test.d.ts

@@ -0,0 +1,4 @@
+export {};
+/**
+ * TODO: [📓] Maybe test all file in examples (not just 10-simple.doc)
+ */

package/esm/typings/src/scrapers/document-legacy/createLegacyDocumentScraper.d.ts

@@ -0,0 +1,21 @@
+import type { ExecutionTools } from '../../execution/ExecutionTools';
+import type { PrepareAndScrapeOptions } from '../../prepare/PrepareAndScrapeOptions';
+import { LegacyDocumentScraper } from './LegacyDocumentScraper';
+/**
+ * Creates a scraper for legacy document formats (.doc, .rtf, etc).
+ * Uses LibreOffice for conversion to extract content from older document formats.
+ *
+ * @public exported from `@promptbook/legacy-documents`
+ */
+export declare const createLegacyDocumentScraper: ((tools: Pick<ExecutionTools, 'llm'>, options: PrepareAndScrapeOptions) => LegacyDocumentScraper) & import("type-fest/source/readonly-deep").ReadonlyObjectDeep<{
+    title: string;
+    packageName: string;
+    className: string;
+    mimeTypes: string[];
+    documentationUrl: "https://github.com/webgptorg/promptbook/discussions/@@";
+    isAvailableInBrowser: false;
+    requiredExecutables: ("Pandoc" | "LibreOffice")[];
+}>;
+/**
+ * TODO: [🎶] Naming "constructor" vs "creator" vs "factory"
+ */

package/esm/typings/src/scrapers/document-legacy/playground/legacy-document-scraper-playground.d.ts

@@ -0,0 +1,5 @@
+#!/usr/bin/env ts-node
+export {};
+/**
+ * Note: [⚫] Code in this file should never be published in any package
+ */

package/esm/typings/src/scrapers/document-legacy/register-constructor.d.ts

@@ -0,0 +1,15 @@
+import type { Registration } from '../../utils/$Register';
+/**
+ * Registration of known scraper
+ *
+ * Warning: This is not useful for the end user, it is just a side effect of the mechanism that handles all available known scrapers
+ *
+ * @public exported from `@promptbook/legacy-documents`
+ * @public exported from `@promptbook/wizard`
+ * @public exported from `@promptbook/cli`
+ */
+export declare const _LegacyDocumentScraperRegistration: Registration;
+/**
+ * TODO: [🎶] Naming "constructor" vs "creator" vs "factory"
+ * Note: [💞] Ignore a discrepancy between file name and entity name
+ */

package/esm/typings/src/scrapers/document-legacy/register-metadata.d.ts

@@ -0,0 +1,28 @@
+import type { Registration } from '../../utils/$Register';
+/**
+ * Metadata of the scraper
+ *
+ * @private within the scraper directory
+ */
+export declare const legacyDocumentScraperMetadata: import("type-fest/source/readonly-deep").ReadonlyObjectDeep<{
+    title: string;
+    packageName: string;
+    className: string;
+    mimeTypes: string[];
+    documentationUrl: "https://github.com/webgptorg/promptbook/discussions/@@";
+    isAvailableInBrowser: false;
+    requiredExecutables: ("Pandoc" | "LibreOffice")[];
+}>;
+/**
+ * Registration of known scraper metadata
+ *
+ * Warning: This is not useful for the end user, it is just a side effect of the mechanism that handles all available known scrapers
+ *
+ * @public exported from `@promptbook/core`
+ * @public exported from `@promptbook/wizard`
+ * @public exported from `@promptbook/cli`
+ */
+export declare const _LegacyDocumentScraperMetadataRegistration: Registration;
+/**
+ * Note: [💞] Ignore a discrepancy between file name and entity name
+ */

package/esm/typings/src/scrapers/markdown/MarkdownScraper.d.ts

@@ -0,0 +1,29 @@
+import type { KnowledgePiecePreparedJson } from '../../pipeline/PipelineJson/KnowledgePieceJson';
+import type { Scraper } from '../_common/Scraper';
+import type { ScraperSourceHandler } from '../_common/Scraper';
+import type { ExecutionTools } from '../../execution/ExecutionTools';
+import type { PrepareAndScrapeOptions } from '../../prepare/PrepareAndScrapeOptions';
+import type { ScraperAndConverterMetadata } from '../_common/register/ScraperAndConverterMetadata';
+/**
+ * Scraper for markdown files
+ *
+ * @see `documentationUrl` for more details
+ * @public exported from `@promptbook/markdown-utils`
+ */
+export declare class MarkdownScraper implements Scraper {
+    private readonly tools;
+    private readonly options;
+    /**
+     * Metadata of the scraper which includes title, mime types, etc.
+     */
+    get metadata(): ScraperAndConverterMetadata;
+    constructor(tools: Pick<ExecutionTools, 'llm'>, options: PrepareAndScrapeOptions);
+    /**
+     * Scrapes the markdown file and returns the knowledge pieces or `null` if it can't scrape it
+     */
+    scrape(source: ScraperSourceHandler): Promise<ReadonlyArray<Omit<KnowledgePiecePreparedJson, 'sources' | 'preparationIds'>> | null>;
+}
+/**
+ * TODO: [🪂] Do it in parallel 11:11
+ * Note: No need to aggregate usage here, it is done by intercepting the llmTools
+ */

package/esm/typings/src/scrapers/markdown/MarkdownScraper.test.d.ts

@@ -0,0 +1,4 @@
+export {};
+/**
+ * TODO: [📓] Maybe test all file in examples (not just 10-simple.md)
+ */

package/esm/typings/src/scrapers/markdown/createMarkdownScraper.d.ts

@@ -0,0 +1,20 @@
+import type { ExecutionTools } from '../../execution/ExecutionTools';
+import type { PrepareAndScrapeOptions } from '../../prepare/PrepareAndScrapeOptions';
+import { MarkdownScraper } from './MarkdownScraper';
+/**
+ * Creates a scraper for markdown content.
+ *
+ * @public exported from `@promptbook/markdown-utils`
+ */
+export declare const createMarkdownScraper: ((tools: Pick<ExecutionTools, 'llm'>, options: PrepareAndScrapeOptions) => MarkdownScraper) & import("type-fest/source/readonly-deep").ReadonlyObjectDeep<{
+    title: string;
+    packageName: string;
+    className: string;
+    mimeTypes: string[];
+    documentationUrl: "https://github.com/webgptorg/promptbook/discussions/@@";
+    isAvailableInBrowser: true;
+    requiredExecutables: never[];
+}>;
+/**
+ * TODO: [🎶] Naming "constructor" vs "creator" vs "factory"
+ */

package/esm/typings/src/scrapers/markdown/playground/markdown-scraper-playground.d.ts

@@ -0,0 +1,5 @@
+#!/usr/bin/env ts-node
+export {};
+/**
+ * Note: [⚫] Code in this file should never be published in any package
+ */

package/esm/typings/src/scrapers/markdown/register-constructor.d.ts

@@ -0,0 +1,15 @@
+import type { Registration } from '../../utils/$Register';
+/**
+ * Registration of known scraper
+ *
+ * Warning: This is not useful for the end user, it is just a side effect of the mechanism that handles all available known scrapers
+ *
+ * @public exported from `@promptbook/markdown-utils`
+ * @public exported from `@promptbook/wizard`
+ * @public exported from `@promptbook/cli`
+ */
+export declare const _MarkdownScraperRegistration: Registration;
+/**
+ * TODO: [🎶] Naming "constructor" vs "creator" vs "factory"
+ * Note: [💞] Ignore a discrepancy between file name and entity name
+ */

package/esm/typings/src/scrapers/markdown/register-metadata.d.ts

@@ -0,0 +1,28 @@
+import type { Registration } from '../../utils/$Register';
+/**
+ * Metadata of the scraper
+ *
+ * @private within the scraper directory
+ */
+export declare const markdownScraperMetadata: import("type-fest/source/readonly-deep").ReadonlyObjectDeep<{
+    title: string;
+    packageName: string;
+    className: string;
+    mimeTypes: string[];
+    documentationUrl: "https://github.com/webgptorg/promptbook/discussions/@@";
+    isAvailableInBrowser: true;
+    requiredExecutables: never[];
+}>;
+/**
+ * Registration of known scraper metadata
+ *
+ * Warning: This is not useful for the end user, it is just a side effect of the mechanism that handles all available known scrapers
+ *
+ * @public exported from `@promptbook/core`
+ * @public exported from `@promptbook/wizard`
+ * @public exported from `@promptbook/cli`
+ */
+export declare const _MarkdownScraperMetadataRegistration: Registration;
+/**
+ * Note: [💞] Ignore a discrepancy between file name and entity name
+ */

package/esm/typings/src/scrapers/markitdown/MarkitdownScraper.d.ts

@@ -0,0 +1,50 @@
+import type { ExecutionTools } from '../../execution/ExecutionTools';
+import type { KnowledgePiecePreparedJson } from '../../pipeline/PipelineJson/KnowledgePieceJson';
+import type { PrepareAndScrapeOptions } from '../../prepare/PrepareAndScrapeOptions';
+import type { Converter } from '../_common/Converter';
+import type { ScraperAndConverterMetadata } from '../_common/register/ScraperAndConverterMetadata';
+import type { Scraper } from '../_common/Scraper';
+import type { ScraperSourceHandler } from '../_common/Scraper';
+import type { ScraperIntermediateSource } from '../_common/ScraperIntermediateSource';
+/**
+ * Integration of Markitdown by Microsoft into Promptbook
+ *
+ * @see https://github.com/microsoft/markitdown
+ * @see `documentationUrl` for more details
+ * @public exported from `@promptbook/markitdown`
+ * @public exported from `@promptbook/pdf`
+ */
+export declare class MarkitdownScraper implements Converter, Scraper {
+    private readonly tools;
+    private readonly options;
+    /**
+     * Metadata of the scraper which includes title, mime types, etc.
+     */
+    get metadata(): ScraperAndConverterMetadata;
+    /**
+     * Markdown scraper is used internally
+     */
+    private readonly markdownScraper;
+    /**
+     * Markdown scraper is used internally
+     */
+    private readonly markitdown;
+    constructor(tools: Pick<ExecutionTools, 'fs' | 'llm' | 'executables'>, options: PrepareAndScrapeOptions);
+    /**
+     * Convert the documents to `.md` file and returns intermediate source
+     *
+     * Note: `$` is used to indicate that this function is not a pure function - it leaves files on the disk and you are responsible for cleaning them by calling `destroy` method of returned object
+     */
+    $convert(source: ScraperSourceHandler): Promise<ScraperIntermediateSource>;
+    /**
+     * Scrapes the source document (PDF, DOCX, etc.) and returns the knowledge pieces or `null` if it can't scrape it.
+     */
+    scrape(source: ScraperSourceHandler): Promise<ReadonlyArray<Omit<KnowledgePiecePreparedJson, 'sources' | 'preparationIds'>> | null>;
+}
+/**
+ * TODO: [🧠][🌜] Export only from `@promptbook/markitdown` or `@promptbook/pdf` NOT both
+ * TODO: [👣] Converted documents can act as cached items - there is no need to run conversion each time
+ * TODO: [🪂] Do it in parallel
+ * Note: No need to aggregate usage here, it is done by intercepting the llmTools
+ * Note: [🟢] Code in this file should never be never released in packages that could be imported into browser environment
+ */

package/esm/typings/src/scrapers/markitdown/createMarkitdownScraper.d.ts

@@ -0,0 +1,22 @@
+import type { ExecutionTools } from '../../execution/ExecutionTools';
+import type { PrepareAndScrapeOptions } from '../../prepare/PrepareAndScrapeOptions';
+import { MarkitdownScraper } from './MarkitdownScraper';
+/**
+ * Constructor of `MarkitdownScraper`
+ *
+ * @public exported from `@promptbook/markitdown`
+ * @public exported from `@promptbook/pdf`
+ */
+export declare const createMarkitdownScraper: ((tools: Pick<ExecutionTools, 'llm'>, options: PrepareAndScrapeOptions) => MarkitdownScraper) & import("type-fest/source/readonly-deep").ReadonlyObjectDeep<{
+    title: string;
+    packageName: string;
+    className: string;
+    mimeTypes: string[];
+    documentationUrl: "https://github.com/webgptorg/promptbook/discussions/@@";
+    isAvailableInBrowser: false;
+    requiredExecutables: never[];
+}>;
+/**
+ * TODO: [🧠][🌜] Export only from `@promptbook/markitdown` or `@promptbook/pdf` NOT both
+ * TODO: [🎶] Naming "constructor" vs "creator" vs "factory"
+ */

package/esm/typings/src/scrapers/markitdown/playground/markitdown-scraper-playground.d.ts

@@ -0,0 +1,5 @@
+#!/usr/bin/env ts-node
+export {};
+/**
+ * Note: [⚫] Code in this file should never be published in any package
+ */

package/esm/typings/src/scrapers/markitdown/register-constructor.d.ts

@@ -0,0 +1,17 @@
+import type { Registration } from '../../utils/$Register';
+/**
+ * Registration of known scraper
+ *
+ * Warning: This is not useful for the end user, it is just a side effect of the mechanism that handles all available known scrapers
+ *
+ * @public exported from `@promptbook/markitdown`
+ * @public exported from `@promptbook/pdf`
+ * @public exported from `@promptbook/wizard`
+ * @public exported from `@promptbook/cli`
+ */
+export declare const _MarkitdownScraperRegistration: Registration;
+/**
+ * TODO: [🧠][🌜] Export only from `@promptbook/markitdown` or `@promptbook/pdf` NOT both
+ * TODO: [🎶] Naming "constructor" vs "creator" vs "factory"
+ * Note: [💞] Ignore a discrepancy between file name and entity name
+ */

package/esm/typings/src/scrapers/markitdown/register-metadata.d.ts

@@ -0,0 +1,28 @@
+import type { Registration } from '../../utils/$Register';
+/**
+ * Metadata of the scraper
+ *
+ * @private within the scraper directory
+ */
+export declare const markitdownScraperMetadata: import("type-fest/source/readonly-deep").ReadonlyObjectDeep<{
+    title: string;
+    packageName: string;
+    className: string;
+    mimeTypes: string[];
+    documentationUrl: "https://github.com/webgptorg/promptbook/discussions/@@";
+    isAvailableInBrowser: false;
+    requiredExecutables: never[];
+}>;
+/**
+ * Registration of known scraper metadata
+ *
+ * Warning: This is not useful for the end user, it is just a side effect of the mechanism that handles all available known scrapers
+ *
+ * @public exported from `@promptbook/core`
+ * @public exported from `@promptbook/wizard`
+ * @public exported from `@promptbook/cli`
+ */
+export declare const _MarkitdownScraperMetadataRegistration: Registration;
+/**
+ * Note: [💞] Ignore a discrepancy between file name and entity name
+ */