# Markdown Toc

> Markdown Toc is an Agent Skill for documents published in the huangruiteng/CS-Notes repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Extract a Markdown heading tree with line numbers before editing or integrating notes. Use when you need to inspect a long Markdown file quickly and choose the best insertion point instead of editing blind.

- Skill name: `markdown-toc`
- Category: [Documents](https://www.openparable.com/skill/documents/1)
- Source: https://github.com/huangruiteng/CS-Notes/blob/HEAD/.codex/skills/markdown-toc/SKILL.md
- Repository: huangruiteng/CS-Notes (3,951 GitHub stars when collected)

## SKILL.md

# Markdown TOC

## Use this skill when

Use this skill before editing a long Markdown file, especially when you need to:
- inspect the heading structure quickly;
- choose the best section for new content;
- avoid creating redundant sections; or
- review nearby lines before patching a note.

## Quick start

Run the bundled script:

```bash
python3 {baseDir}/scripts/extract_toc.py /absolute/path/to/file.md
```

In this repository, common targets include:

```bash
python3 {baseDir}/scripts/extract_toc.py Notes/Gourmet.md
python3 {baseDir}/scripts/extract_toc.py Notes/AI-Agent-Product&PE.md
```

## Workflow

1. Run the TOC script on the candidate file.
2. Read the returned heading tree and line numbers.
3. Open the most likely section neighborhood with `sed -n '<start>,<end>p'` or `nl -ba`.
4. Insert into an existing section when possible.
5. Create a new section only when the file truly lacks a clean landing spot.

## Output format

The script prints a compact tree like this:

```text
TOC
--------------------------------------------------
# Title (line 1)
  ## Section (line 20)
    ### Subsection (line 42)
```

---

Source: https://www.openparable.com/skill/documents/markdown-toc (OpenParable). Cite the canonical URL above.