llms.txt

llms.txt gives Large Language Models and AI agents a clean, curated, machine-first map of a site’s content — a table of contents written for inference, not browsers. It sits beside its cognitive-filesystem siblings SKILL.md, AGENTS.md, and the lingo glossary.

---

1. 🧬 Anatomy

Canonical skeleton of a compliant llms.txt, top to bottom — clean Markdown in a fixed order.

# Site/Project Name
 
> Blockquote containing a short, descriptive context summary.
 
Detailed project descriptions and usage guidelines go here.
 
## 2. Navigation Group Heading
 
- [Link Name](https://ssot.0rk.de/target): Optional description text separated by a colon.
- [Additional Link](/local-path): Description details.

Part	Required	Rule
H1 Header	✅	Project name on the very first line.
Blockquote	✅	One-paragraph summary of the site’s context.
Body	optional	Free-form prose for usage guidelines and caveats.
H2 Navigation	recommended	Group links logically under section headings.
Links	recommended	Standard [Name](URL) followed optionally by : Description.

---

3. 🚀 Deployment

Serve raw text files from the domain root.

• Location: Host at the root of the target domain:
  • https://ssot.0rk.de/llms.txt — streamlined navigation and high-level table of contents.
  • https://ssot.0rk.de/llms-full.txt — optional full-text compilation of all target documents.
• Static Serving: Drop the raw text files directly into the public or content directories of static site generators (Quartz, Docusaurus, VitePress).

---

4. 📦 Tooling

• WordPress: integrate via SEO plugins (e.g. Yoast, AIOSEO).
• CI/CD: generate dynamically during builds with project crawlers or custom scripts.

---

5. 💡 Best Practices

• Jargon-Free Language: keep descriptions simple and direct for optimal LLM context parsing.
• Clean Markdown: avoid complex HTML tags or styles so scrapers stay compatible.
• Unambiguous Paths: use absolute URLs or root-relative paths for reliable link resolution.

---

6. 🚨 Gotchas

• Not an Official Protocol: major AI platforms do not commit to parsing llms.txt automatically.
• Complementary to Robots: llms.txt guides attention but never replaces the permission controls in robots.txt.
• Token Optimization: keep llms.txt high-level and offload full text to llms-full.txt to minimize token overhead.