md-to-toc is a python script that parses a markdown file and outputs a table of contents as a markdown list.

Advantages:

• Doesn't require an internet connection
• Works on any markdown file (doesn't need to be on a public GitHub repo)
• Fast

Disadvantages:

• Uses my best guess at GitHub's algorithm to determine anchor names for header strings. Works for my use cases, but may not cover all cases. If it doesn't work for you, please let me know!

• Install Python 2.7 (haven't tested with other versions)
• Run md-to-toc <markdown_file>
• Copy output to your markdown file

Given test.md:

# Header1
Some text

## Header2
Some more text

## Header2 again
Even more text

### Header3

#### Header4

##### Header5

###### Header6

# Header1

## Header2

## Header2 again

Run md-to-toc:

$ python md-to-toc.py test.md
- [Header1](#header1)
  - [Header2](#header2)
  - [Header2 again](#header2-again)
    - [Header3](#header3)
      - [Header4](#header4)
        - [Header5](#header5)
          - [Header6](#header6)
- [Header1](#header1-1)
  - [Header2](#header2-1)
  - [Header2 again](#header2-again-1)

One cute feature is that if your file doesn't start with header level 1, leading spaces will be removed so that the output is flush with the left edge:

test2.md:

### Header3

#### Header4

##### Header5

###### Header6

$ python md-to-toc.py test2.md
- [Header3](#header3)
  - [Header4](#header4)
    - [Header5](#header5)
      - [Header6](#header6)

github-markdown-toc is a bash script that uses GitHub's API to convert a markdown file to HTML, then parses the HTML to produce the table of contents. This script is more robust than md-to-toc because it parses out the anchors directly from HTML, while md-to-toc converts the header titles to anchors using my best guess at GitHub's algorithm. On the other hand, github-markdown-toc requires an internet connection.