Add markdown-split for context loading#9
Open
coolai86 wants to merge 6 commits into
Open
Conversation
added 6 commits
April 5, 2026 00:07
Single-binary Go web application patterns: - Project structure (cmd/, internal/, web/dist/) - Single embed file pattern - SPA routing with fallback handler - Development mode patterns - Anti-patterns to avoid Framework-agnostic: no vite/esbuild references.
- Document sql.Scanner/driver.Valuer pattern for custom types - Add JSONB override configuration example - Add nullable types guidance (sql.Null[T] and legacy types) - Update PostgreSQL type mappings table to reference custom types This helps projects use sqlc effectively for JSONB columns, domain-specific types, and custom nullable types.
- download-markdown.cjs: Download URL, detect Markdown via extensions, Accept headers, and HTML link parsing - html-to-markdown-recursive.cjs: Convert HTML files to Markdown recursively - website-download-recursive.sh: wget-based recursive website download - Markdown detection: 8 strategies including URL variants, Accept headers, and HTML link parsing (edit/source buttons, .md links, raw repo URLs) - Dependencies: @mozilla/readability, turndown, jsdom - Supports content-type headers, content sniffing, and frontmatter generation
- New script to split large markdown files for context loading - View TOC with line numbers (--toc) - Split by headers with size limits - Create index file for navigation - Skip code blocks and YAML frontmatter when parsing - Handles oversized sections with warnings Enables progressive disclosure: grep for headers, read only needed sections.
Modern Node.js supports 'import' on CJS modules natively. - Renamed .cjs → .mjs - Converted require() to import statements - Updated SKILL.md examples
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Adds
markdown-split.cjsto the download-markdown skill for splitting large markdown files into smaller reference files.Problem: Large markdown docs exceed context windows. Reading 100KB of documentation burns tokens.
Solution: Progressive disclosure — split by headers, read only what you need.
Features:
--toc— View TOC with line numbers before splitting--index— Create INDEX.md listing all splits--depth N— Header depth to split on (default: 2)--max-lines/--max-bytes— Size limits per file--flatten— Flat numbered filenamesWorkflow:
Creates files suitable for progressive disclosure: use
grep -n '^#' file.mdto find headers, then read specific sections.