@packmd/core
v1.0.7
Published
Core parsing and markdown generation logic for PackMD
Maintainers
Readme
@packmd/core
Shared core logic for PackMD – the engine that powers both the web app and the CLI.
This package handles everything from fetching repositories and scraping web pages to building directory trees and generating clean, token‑efficient Markdown.
Features
- GitHub fetching – retrieve repository metadata, list files, download content (supports private repos with token).
- Web scraping – uses the Jina Reader API to convert any web page into clean Markdown.
- Local directory scanning – recursively walk directories with
.gitignoresupport. - Smart filtering – apply glob patterns, size limits, binary detection, and ignore rules.
- Tree building – construct a hierarchical file tree from flat file lists.
- Markdown generation – produce a consistent, LLM‑ready output with headers, stats, and code blocks.
Installation
npm install @packmd/core
# or
bun add @packmd/coreNote: This package is intended for internal use by the PackMD ecosystem. You typically don’t need to install it directly unless you are extending or contributing to the project.
Main Exports
| Export | Description |
| -------------------- | --------------------------------------------------------------------- |
| fetchGithubRepo | Fetch a GitHub repository and return a Markdown digest. |
| scrapeWebPage | Scrape a web page via Jina Reader and return its content as Markdown. |
| buildFileTree | Build a nested FileNode tree from an array of RawFile objects. |
| generateMarkdown | Generate a complete Markdown string from a FileNode tree. |
| buildDigestHeader | Build a consistent header section for the digest. |
| matchesAny | Test a path against an array of glob patterns. |
| DEFAULT_IGNORES | Default exclude patterns for local and GitHub repositories. |
| packmdSchema (Zod) | Schema for validating user input (URL, options, globs, etc.). |
Basic Usage Example
import { fetchGithubRepo, generateMarkdown, buildFileTree } from "@packmd/core"
// Fetch a public GitHub repo
const result = await fetchGithubRepo("https://github.com/vercel/next.js", {
maxFiles: 100,
maxFileSizeKB: 50,
excludeGlobs: ["**/*.test.ts"],
})
// `result.markdown` contains the complete digest
console.log(result.markdown)
// If you want to build a tree manually:
const tree = buildFileTree(result.files)
const markdown = generateMarkdown(tree, { showTree: true })License
MIT © Holiday
