Token Optimizer

Compress prompts via a plain REST endpoint.

POST /api/v1/compress-prompt compresses a prompt to use fewer tokens while preserving every instruction and constraint, verified against a real tokenizer — not an estimate.

Request

bash

curl -X POST https://cuelara.com/api/v1/compress-prompt \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_TOKEN_HERE" \
  -d '{
    "text": "Could you please help me write a Python script that...",
    "level": "Aggressive (Max Savings)"
  }'

The Authorization header is optional — omit it to call anonymously (lower rate limit). Get a token from /dashboard/mcp.

Body

text (string, required) — the prompt to compress.
level (string, optional) — "Low (Safest)", "Medium (Balanced)" (default), or "Aggressive (Max Savings)".
preserveFormatting (string, optional) — "Yes" (default) or "No".

Response

json

{
  "compressed": "Compress this prompt. Keep all instructions and constraints intact.",
  "originalTokens": 61,
  "compressedTokens": 12,
  "savedPercent": 80
}

Errors

json

400 invalid body, 401 invalid token, 429 rate limit, 502/503/504 upstream AI failure.

Limits

Anonymous calls share the same free daily limit as the website's Token Optimizer, keyed by IP. Signed-in calls use your own account's plan limits instead. See the API overview for auth and CORS details.