5 Text Tools Every AI Agent Needs: Stats, Embeddings, Markdown, and More with IteraTools

If you're building AI agents or LLM-powered apps, you've probably run into the same set of recurring needs: analyze text, generate embeddings for RAG, convert markdown to HTML, summarize long documents, send formatted output somewhere. These are the small, utility-type operations that every AI pipeline eventually needs — and they shouldn't require spinning up a separate service for each one.

This post covers 5 text processing tools now available in IteraTools — a pay-per-use API for AI agents built on the x402 micropayment protocol (Base/USDC). No subscriptions, no quotas, just $0.001–$0.002 per call.

---

1. POST /text/stats — Text Statistics Analysis

Price: $0.001/request

The simplest one. Send any text, get back a full breakdown:

curl -X POST https://api.iteratools.com/text/stats \
  -H "Authorization: Bearer YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"text": "The quick brown fox jumps over the lazy dog. Pack my box with five dozen liquor jugs."}'

Response:

{
  "words": 16,
  "characters": 89,
  "characters_no_spaces": 74,
  "sentences": 2,
  "paragraphs": 1,
  "unique_words": 16,
  "avg_word_length": 4.6,
  "reading_time_seconds": 4,
  "reading_time_label": "less than a minute"
}

Useful for: enforcing content length limits, estimating reading time, validating that generated text meets certain criteria before sending it downstream.

Runs entirely locally — no external API calls, near-instant response.

---

2. POST /embeddings — Vector Embeddings (OpenAI)

Price: $0.001/request

Generate embeddings using OpenAI's text-embedding-3-small (1536 dimensions) or text-embedding-3-large (3072 dimensions). Supports batches of up to 100 strings.

curl -X POST https://api.iteratools.com/embeddings \
  -H "Authorization: Bearer YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"text": ["How do I reset my password?", "Password reset instructions"], "model": "small"}'

Response:

{
  "ok": true,
  "data": {
    "embeddings": [[0.023, -0.041, ...], [0.019, -0.038, ...]],
    "model": "text-embedding-3-small",
    "tokens": 12,
    "dimensions": 1536,
    "count": 2
  }
}

Use cases: semantic search, RAG pipelines, clustering, deduplication, recommendation systems. At $0.001/request for batches of up to 100, it's extremely cost-effective for production workloads.

---

3. POST /markdown/render — Markdown to HTML

Price: $0.001/request

Convert Markdown to HTML using the marked library (CommonMark compliant). Runs locally, no external API.

curl -X POST https://api.iteratools.com/markdown/render \
  -H "Authorization: Bearer YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"markdown": "# Hello\n\nThis is **bold** text with a [link](https://iteratools.com)."}'

Response:

{
  "html": "<h1>Hello</h1>\n<p>This is <strong>bold</strong> text with a <a href=\"https://iteratools.com\">link</a>.</p>\n",
  "input_length": 67,
  "output_length": 102
}

Supports: headings, bold/italic, lists, links, code blocks (inline and fenced), blockquotes, tables, and horizontal rules.

This is particularly useful for AI agents that generate Markdown as output (most LLMs do by default) but need to deliver HTML — for email templates, CMS content, documentation sites, or UI rendering.

---

4. POST /summarize — Extractive Text Summarization

Price: $0.002/request

Extract the most important sentences from any text or URL using TF-IDF scoring. No LLM involved — pure extractive summarization, fast and deterministic.

curl -X POST https://api.iteratools.com/summarize \
  -H "Authorization: Bearer YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"url": "https://en.wikipedia.org/wiki/Artificial_intelligence", "sentences": 3}'

Response:

{
  "summary": "Artificial intelligence (AI) is the simulation of human intelligence processes by machines, especially computer systems...",
  "sentences": 3,
  "original_length": 45210,
  "compression_ratio": 0.018
}

You can also pass text directly instead of a URL. Accepts up to 50,000 characters. Great for preprocessing long documents before sending them to an LLM — reduces tokens and cost.

---

5. POST /sentiment — Sentiment Analysis

Price: $0.001/request

Analyze text sentiment using the AFINN-165 lexicon. Returns positive/negative/neutral classification with a numeric score.

curl -X POST https://api.iteratools.com/sentiment \
  -H "Authorization: Bearer YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"text": "I absolutely love this product! It works great and the support team is fantastic."}'

Response:

{
  "sentiment": "positive",
  "score": 9,
  "comparative": 1.125,
  "positive": ["love", "great", "fantastic"],
  "negative": []
}

Fast, local, no external calls. Perfect for monitoring, content moderation pre-filtering, or routing user messages to different handlers based on emotional tone.

---

Using with MCP (Claude, Cursor, etc.)

All of these tools are available in the mcp-iteratools package:

npx mcp-iteratools

Or in your MCP config:

{
  "mcpServers": {
    "iteratools": {
      "command": "npx",
      "args": ["mcp-iteratools"],
      "env": {
        "ITERATOOLS_API_KEY": "your-key-here"
      }
    }
  }
}

Once connected, Claude/Cursor can use markdown_render, embeddings, text_stats, summarize, and sentiment directly as tools — no additional setup required.

---

Pricing Model

IteraTools uses x402 — a micropayment protocol built on Base (USDC). You pay per request, no monthly fees, no quotas.

• Get an API key at iteratools.com
• Top up with USDC on Base
• Pay $0.001–$0.002 per text tool call

For the 5 tools above, processing 10,000 requests costs between $10 and $20. Compare that to hosted LLM APIs for the same volume.

---

What's Next

The roadmap includes more text tools: keyword extraction, language detection, readability scores, and HTML-to-Markdown (the reverse of today's endpoint).

Full API docs at iteratools.com/docs · All tools · npm package