llms.txt Analyzer & Previewer
Analyze the structure and content of an llms.txt file. Preview the rendered Markdown, inspect sections and links, and calculate words, characters, file size and estimated tokens.
Select Input Method
Project Title
Context Estimate
Tokens are estimated using a generic ~4 chars/token heuristic. Actual counts depend on the specific LLM tokenizer.
Observations
Rendered Markdown Preview
Structural Outline
Section Distribution
| Section Name | Links | Described | Internal | External | Tokens |
|---|
Resource Links
| Title | URL | Section | Type |
|---|
Detailed File Statistics
Basic Metrics
- Total Words0
- Characters (no spaces)0
- Characters (with spaces)0
- Total Lines0
- Blank Lines0
- File Size0 KB
Link Metrics
- Total Links0
- Unique URLs0
- Duplicate URL Occurrences0
- Internal Links0
- External Links0
- Unique Domains0
What Is the llms.txt Analyzer?
The llms.txt Analyzer & Previewer is a read-only diagnostic dashboard designed to help you understand exactly what your llms.txt file contains. Instead of manually counting URLs or guessing token sizes, this tool instantly parses your file, extracts all sections and links, and builds a comprehensive statistical report.
How the Analyzer Works
When you paste or upload your file, the Analyzer runs a specialized Markdown parser entirely in your browser. It identifies the H1 project title, any blockquote summaries, introductory text, and H2 file-list sections. It then extracts every Markdown link, categorizes them as internal or external, measures their descriptions, and calculates size metrics for the entire document.
What Information the Tool Measures
The tool provides an exhaustive breakdown including total words, characters, physical lines, unique URLs, and duplicate occurrences. It also visualizes the distribution of content across your sections, making it easy to see if a single section holds 90% of your links or if your descriptions are missing.
How Estimated Token Counting Works
Because different Large Language Models (like GPT-4, Claude, or Llama) use different tokenization algorithms, it is impossible to give a single perfect token count. The Analyzer uses a standard heuristic (approximately 4 characters per token) to provide a solid baseline estimate, helping you plan your context window budgets safely.
Core vs Optional Preview
The official specification defines a special ## Optional section for secondary resources (like blogs or privacy policies). The Analyzer automatically separates this section and calculates two different token estimates: a "Core File" context and a "Full File Context." This shows you exactly how much weight your optional resources add to the payload.
Analyzer vs Validator
The Analyzer is designed to measure and visualize your file. While it may point out basic structural observations (like missing summaries or duplicate URLs), it does not enforce strict format rules. To ensure your file complies with all official recommendations, you should run it through the llms.txt Validator.