Skip to content

System Prompt Analyzer

Estimate a system prompt's tokens for Claude, GPT, Llama and Gemini, its share of a 4K to 200K context window, and its template variables.

System Prompt Analyzer

Start with a template

Stats

Characters

436

Words

62

Lines

13

Estimated tokens (claude)

125

Context limits

4K4,096
8K8,192
16K16,384
32K32,768
65K65,536
128K128,000
200K200,000

Variables detected

retrieved_documents1×
user_query1×
Pro tips: Use XML tags like <system>, <context>, and <instructions> to structure your prompt. Define variables with {{variable}} syntax.

Updated . Provided as is. Check the output before you rely on it in production.

How to use System Prompt Analyzer

  1. 1

    Paste or start from a template

    Paste your system prompt or start from a template such as General Assistant, Code Review or RAG System.

  2. 2

    Check size against the context

    Read characters, words, lines and estimated tokens for Claude, GPT, Llama or Gemini, against common context limits.

  3. 3

    Review structure

    Detected template variables and XML tags show how the prompt is organized. Wrap selections in tags to separate instructions from context.

  4. 4

    Copy or download

    Copy the prompt or download it as a .txt file.

Questions and answers

What is System Prompt Analyzer?
A system prompt is the standing instruction text a model reads before every user message, and its tokens are paid on every request. This analyzer estimates them per model family, shows the share of a 4K to 200K context window, counts words and lines, and lists template variables. It uses System Prompt Editor's editor.
Which models does it estimate tokens for?
Claude, GPT, Llama and Gemini, using character-ratio estimates per model family.
Is my prompt uploaded?
No. Analysis runs entirely in your browser.
For AI agents: how to call this tool

Machine-readable contract, endpoints and examples. Humans can ignore this section.

Best Path For Builders

Browser workflow

Runs instantly in the browser with private local processing and copy/export-ready output.

Browser Workflow

This tool is optimized for instant in-browser execution with local data handling. Run it here and copy/export the output directly.

/system-prompt-analyzer/

For automation planning, fetch the canonical contract at /api/tool/system-prompt-analyzer.json.