Text Analysis API: Sentiment, Keywords, Readability in One Call

Sentiment analysis, keyword extraction, and readability scoring are three separate libraries you don't need to install. Here's how to get all three in one API call.

The Single-Call Approach

import requests

text = """
Cloudflare Workers is an incredible platform for building APIs.
The free tier is genuinely useful for production workloads,
though the 10ms CPU limit can be frustrating for compute-heavy tasks.
"""

resp = requests.post("https://api.lazy-mac.com/text-analysis/analyze", json={
    "text": text,
    "features": ["sentiment", "keywords", "readability", "language"]
})

analysis = resp.json()

Response:

{
  "sentiment": {
    "label": "mixed",
    "score": 0.62,
    "positive": 0.71,
    "negative": 0.29
  },
  "keywords": [
    {"word": "Cloudflare Workers", "score": 0.94},
    {"word": "free tier", "score": 0.87},
    {"word": "10ms CPU limit", "score": 0.71}
  ],
  "readability": {
    "flesch_score": 58.3,
    "grade_level": "10th grade",
    "avg_sentence_length": 18.2,
    "complex_word_ratio": 0.14
  },
  "language": "en",
  "word_count": 42
}

Use Cases

Content moderation

def is_safe_content(text: str) -> bool:
    analysis = analyze(text)
    return analysis['sentiment']['negative'] < 0.7

SEO optimization

def optimize_for_seo(text: str) -> list:
    analysis = analyze(text)
    keywords = [k['word'] for k in analysis['keywords'][:5]]
    if analysis['readability']['grade_level'] > '12th grade':
        return {'action': 'simplify', 'target_keywords': keywords}
    return {'action': 'publish', 'keywords': keywords}

Customer feedback triage

feedback_items = get_support_tickets()
for ticket in feedback_items:
    analysis = analyze(ticket['text'])
    if analysis['sentiment']['negative'] > 0.8:
        escalate_to_human(ticket)
    else:
        send_auto_reply(ticket, analysis['keywords'])

Text Analysis API | Full API store