原始内容
markdown-reader
An Agent Skill for progressively reading Markdown. Get the outline first, then extract only the sections you need.
✅ What it does
- 🧭 Get a heading outline
- ✂️ Extract only the needed sections
- 🧩 Read step by step for better understanding
📥 Expected input
- README.md
- Markdown specs or guides
📤 Output (summary)
- Heading outline
- Extracted section content
💡 Example prompts
After setting up the skill in your agent, use prompts like:
- "Read this Markdown with markdown-reader"
- "Show the outline first, then read this section"
You can load it incrementally like this:

📦 Prerequisite - Python 3
Python 3 is required. This skill ships with a lightweight treemd-compatible script.
Use the script from the skill's scripts/ directory.
📦 Install this skill
git clone https://github.com/Tomatio13/markdown-reader-skill.git
cd markdown-reader-skill
cp -pr markdown-reader ~/.codex/skills
Copy it into your agent's skill directory.
⚠️ Notes
--sectionrequires an exact heading match- If a heading is not found, check spelling and hierarchy
🗂️ Files
markdown-reader/SKILL.mdmarkdown-reader/scripts/md-index.pymarkdown-reader/references/md-index.md
Acknowledgements
- Inspired by treemd for compatible CLI behavior.