@agentutility/mcp-wordmint

Named LLM tasks. Minted at a sticker price.

Summarize, translate, classify, extract entities, generate regex, score a resume — every named text task an agent needs, with a fixed price and no prompt engineering.

Pricing: pay-per-call in USDC on Base. No subscriptions, no API keys. See per-tool prices below.

Install — Claude Desktop

Edit ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

{
  "mcpServers": {
    "agentutility-wordmint": {
      "command": "npx",
      "args": ["-y", "@agentutility/mcp-wordmint"],
      "env": { "X402_PRIVATE_KEY": "0xYOUR_PRIVATE_KEY_HEX" }
    }
  }
}

Restart Claude Desktop. 114 tools appear in the tool palette.

Install — Cursor

Add to .cursor/mcp.json:

{
  "mcpServers": {
    "agentutility-wordmint": {
      "command": "npx",
      "args": ["-y", "@agentutility/mcp-wordmint"],
      "env": { "X402_PRIVATE_KEY": "0x..." }
    }
  }
}

Funding

Send any amount of USDC on Base mainnet to the address derived from your X402_PRIVATE_KEY. The MCP server uses it to pay for tool calls automatically.

USDC on Base contract: 0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913

Tools (114)

| Tool | Description | |---|---| | ai-to-human-text | (0.01 USDC/call) Rewrite AI-generated text so it reads like a person wrote it. Send text plus an optional tone (neutral, casual, professional, or academic) and it strips formulaic transitions, hedging phrases, corporate filler words, and overused em-dashes while preserving meaning and, by default, the original length within about 15%. Returns the rewritten text along with input and output character counts. Use it as an AI text humanizer, GPT detector bypass, or AI content rewriter before publishing drafted copy. | | alt-text-generator | (0.02 USDC/call) Turns a public image URL into ready-to-use text: concise alt text for screen readers, a natural-language description, OCR text pulled from the image, a list of tags, or a one-line social caption. Send image_url plus an optional mode (describe, alt_text, ocr, tags, or caption) and an optional custom prompt to override the instruction; get back text, the mode used, and the model that generated it. Use it as an alt text generator, image captioning API, image-to-text OCR tool, or accessibility description service for photos, screenshots, and graphics. | | app-review-sentiment | (0.01 USDC/call) Scores app-store reviews by onboarding, stability, pricing, performance, and feature requests. Same sentiment-analysis backend. Use it as an app review sentiment API, mobile app feedback classifier, or store review analyzer. | | brand-bootstrap | (0.01 USDC/call) Bootstraps a brand kit for a new business or product in one call. Same backend as brand-launch-brief, exposed under a workflow slug for agents bootstrapping a company or product. Returns name score, positioning, tagline options, messaging pillars, logo prompt, launch checklist, and suggested channels. Use it as a brand bootstrap API or new business launch brief. | | brand-launch-brief | (0.01 USDC/call) Generates a structured brand launch brief for a new company or product from name, concept, audience, and tone. Returns name score, positioning, tagline options, messaging pillars, logo prompt, launch checklist, and suggested launch channels. Designed to compose with domain-suggest, domain-availability, x-handle-availability, logo generation, and brand-clearance. Use it as a brand launch brief generator, startup bootstrap brief, new company brand kit, or product positioning brief. | | brand-positioning-brief | (0.01 USDC/call) Generates a brand positioning brief covering messaging pillars, taglines, launch channels, and a logo prompt. Same backend as brand-launch-brief. Use it as a company positioning generator or product messaging brief. | | brand-sentiment | (0.03 USDC/call) Measures brand sentiment on X by searching recent public posts about a brand, product, token, or launch and returning an aggregate positive/neutral/negative distribution plus tweet-level labels. Same read-only X sentiment backend as twitter-sentiment. Use it as a brand sentiment API, social listening sentiment tool, or X brand monitor. | | brand-sentiment-analysis | (0.01 USDC/call) Scores comments, mentions, survey verbatims, and campaign feedback for sentiment. Same backend as sentiment-analysis. Use it as a brand sentiment analysis API, reputation monitoring tool, or campaign sentiment classifier. | | brand-tagline | (0.005 USDC/call) Generates brand taglines and slogans for launch pages, X bios, email copy, and product cards. Same backend as brand-tagline-generate under a shorter discovery slug. Use it as a brand tagline API, company slogan generator, or product positioning line generator. | | brand-tagline-generate | (0.005 USDC/call) Generates tagline options for a brand or startup from its name, concept, audience, and tone. Useful for agents bootstrapping landing pages, X bios, launch emails, and brand briefs. Use it as a brand tagline generator, startup tagline API, or company positioning line generator. | | candidate-feedback-sentiment | (0.01 USDC/call) Summarizes candidate experience, recruiter feedback, and hiring-process comments by sentiment. Same backend as sentiment-analysis. Use it as a candidate feedback sentiment API, recruiting survey classifier, or interview feedback emotion scorer. | | card-resolve | (0.005 USDC/call) Normalizes free-form graded card text into a canonical card object. Takes any descriptive string ('2018 Topps Update Juan Soto RC PSA 9', '1999 Pokemon Base Set Charizard Holo PSA 10', 'MTG Black Lotus Alpha BGS 8.5') and returns sport_or_tcg, year, set, player_or_card_name, parallel, grade, grader, and cert_number, plus a confidence score and a stable search_key suitable for downstream cache coherence. Structured extraction from the text you provide, with no external data lookups. The clean glue endpoint for collector-pricing bots, marketplace listing automations, AI-shopping agents, and portfolio trackers that need to normalize messy human descriptions before any further lookup. Use it as a card resolver or graded card string normalizer. | | churn-risk-sentiment | (0.01 USDC/call) Analyzes support tickets, chats, and emails for negative sentiment, urgency, and at-risk churn themes. Same backend as sentiment-analysis. Use it as a churn risk sentiment API, customer frustration classifier, or retention signal scorer. | | citation-verify | (0.01 USDC/call) Verifies whether a claim is actually supported by a source URL. Fetches the source (HTML stripped, 100KB cap) and asks Venice claude-sonnet-4-6 whether the claim is supported, contradicted, or absent. Returns verdict, confidence (0-1), and the strongest evidence quote when available. Use it as a citation verifier, fact-check against URL, does-this-source-support-this-claim check, or hallucination detector. | | classify | (0.02 USDC/call) Sort text into categories you define on the spot, no training run required. Send text and a labels array (2-25 options), with optional label_descriptions for extra context and a multi_label flag for cases where more than one tag applies, and get back a results array with calibrated confidence scores, a top_label, and a one-sentence reasoning string explaining the call. Confidence is honest rather than inflated: ambiguous text gets scored lower instead of forced into a category. Use it as a zero-shot text classifier, content moderation filter, or intent/topic tagging API for routing tickets, labeling feedback, or sorting incoming messages. | | classify-text | (0.02 USDC/call) Sorts a piece of text into categories you define on the fly, no training or fixed label set required. Send text plus labels (2 to 25 caller-supplied categories), an optional multi_label flag to allow more than one match, and optional label_descriptions to disambiguate close labels; get back results with a calibrated confidence per label, a top_label, and a one-sentence reasoning for the call. Use it as a zero-shot text classifier, text classification API, content moderation router, or category and label sorter for tickets, reviews, and inbound messages. | | commit-message-from-diff | (0.01 USDC/call) Generates a git commit message from a diff in Conventional Commits format. Returns type, scope, subject, body, and a breaking_change flag, ready for git commit -m. Use it as an AI commit-msg generator. | | community-moderation-sentiment | (0.01 USDC/call) Scores community posts, replies, and threads by tone, frustration, escalation risk, and constructive intent. Same sentiment-analysis backend. Use it as a community moderation sentiment API, forum tone classifier, or community health signal. | | content-simhash | (0.005 USDC/call) Fingerprints text with a 64-bit SimHash for near-duplicate detection, computed entirely locally. Uses token-level k-shingles (default k=3) with FNV-1a; two SimHashes are 'close' (small Hamming distance) iff the underlying texts share many shingles. Returns hex + decimal forms plus token + shingle counts. Useful for content dedup pipelines, plagiarism detection, and bot-content clustering. Use it as a content fingerprint, dedup hash, or locality-sensitive hash. | | contract-translate | (0.002 USDC/call) Translates contract excerpts and commercial terms, preserving numbering, defined terms, clause references, dates, and party names. Not legal advice. Same backend as translate-text. Use it as a contract translation API, agreement translator, or legal document localization tool. | | cron-explain | (0.002 USDC/call) Cron expression parser. Takes a 5/6/7-field cron expression or @yearly/@monthly/@weekly/@daily/@hourly macro and returns a plain-English explanation, frequency bucket, validation result, and a runs-per-day estimate. Useful for explaining a schedule to a human, validating a cron string, or generating a cadence summary. | | cron-next | (0.003 USDC/call) Tells you exactly when a cron expression fires next. Takes any 5/6/7-field cron expression (or @daily/@hourly/@weekly/@monthly/@yearly macros) plus a timezone and returns the next N firing times as ISO 8601, Unix epoch seconds, and a human-readable string in the requested timezone. Also reports whether the cadence is regular (constant interval between fires) and the interval in seconds when it is. Companion to cron-explain, which describes the cadence in English. Use it as a cron next-fire-times forecaster, schedule preview, or cadence calculator. | | cron-parse | (0.005 USDC/call) Cron parser. Parses any 5 / 6 / 7-field cron expression or @yearly / @monthly / @weekly / @daily / @hourly macro and returns a plain-English explanation, frequency bucket, validation flag, and runs-per-day estimate. Same backend as cron-explain under a clearer slug. | | crypto-news-sentiment | (0.01 USDC/call) Classifies supplied crypto news headlines, project updates, or community posts by sentiment and emotion. Classification only, not investment advice. Same sentiment-analysis backend. Use it as a crypto news sentiment API, token headline sentiment tool, or Web3 community tone scorer. | | crypto-sentiment | (0.03 USDC/call) Scores crypto sentiment from recent public X posts about protocols, tickers, tokens, x402, Bankr, Coinbase, and market narratives, without posting or account mutation. Same read-only X sentiment backend as twitter-sentiment. Use it as a crypto sentiment API, token narrative sentiment tool, or X market sentiment search. | | customer-review-translate | (0.002 USDC/call) Translates customer reviews, ratings comments, and voice-of-customer excerpts, preserving product names, SKUs, and star ratings. Same translate-text backend. Use it as a customer review translation API, marketplace review localization tool, or feedback translator. | | describe-image | (0.02 USDC/call) Describes images with a vision LLM across five modes: describe, alt_text (accessibility, <=125 chars), OCR (extract visible text), tags (8-15 keywords), and caption (single-sentence). Use it as an AI image descriptor or describe-image endpoint. | | detect-language | (0.005 USDC/call) Identifies what language a piece of text is written in and how confident the call is, returning an ISO 639-1 code, the full language name, and a 0-1 confidence score. Send text and get back language_code, language_name, confidence, plus is_mixed and secondary_languages when more than one language is present in meaningful proportion. Handles short snippets and mixed-language input across a wide range of languages. Use it as a language detector, language identification API, or language identifier for routing multilingual content, tagging support tickets, or filtering text pipelines by language. | | detect-pii | (0.02 USDC/call) Detects PII in text: emails, phones, SSNs, credit cards, addresses, names, IPs, and API tokens. Returns matches + risk_level, with optional redaction using [TYPE] tokens. Use it as a PII detector, personal data scanner, data leak scanner, or privacy redaction API. | | dictionary | (0.005 USDC/call) Dictionary API for looking up an English word and getting its definitions, part of speech, pronunciation, usage examples, and per-sense synonyms. Send one word and receive grouped meanings, IPA phonetics, and a pronunciation audio URL when available. Use it as a dictionary API, word definition lookup, English word reference, or pronunciation lookup. | | dictionary-define | (0.005 USDC/call) Looks up English word definitions with pronunciation, part of speech, and synonyms. Returns part-of-speech-grouped definitions with usage examples and per-sense synonyms, plus IPA phonetic and a pronunciation audio URL when available. Wraps the public Free Dictionary API, no auth, commercial-OK. Use it as an English dictionary or etymology-adjacent word lookup. | | duplicate-ticket-detect | (0.008 USDC/call) Checks a new support ticket against a batch of existing ones and ranks which are likely the same issue, so you can merge or link duplicates before they get worked twice. Send the new ticket plus up to 500 candidate tickets ({id?, text}) and get back matches sorted by lexical similarity score, each with the candidate index, id, score, and a text snippet, a deterministic word-shingle Jaccard comparison with no LLM call. Built for duplicate ticket detection, similar tickets, and support dedupe workflows; redact customer PII from ticket text upstream before sending it. Use it as a duplicate ticket detection API, similar-ticket finder, support dedupe tool, or ticket matching endpoint ahead of manual triage. | | ecommerce-review-sentiment | (0.01 USDC/call) Scores ecommerce reviews by product aspect, emotion, and escalation risk. Same backend as sentiment-analysis. Use it as an ecommerce review sentiment API, marketplace review classifier, or product feedback scorer. | | education-course-translate | (0.002 USDC/call) Translates course outlines, assignments, quiz instructions, and LMS copy, preserving placeholders and numbering. Same translate-text backend. Use it as an education course translation API, lesson localization tool, or training content translator. | | email-draft | (0.02 USDC/call) Writes emails with AI: subject, body, salutation, and sign-off. 7 tones x 10 email types (cold_outreach, follow_up, decline, reply, internal_update, thank_you, intro, request, apology, general), with length controls. Use it as an AI email writer, cold outreach drafter, or follow-up generator. | | embedding-similarity | (0.005 USDC/call) Measures how semantically similar two strings are: embeds both via Venice (default model: text-embedding-bge-m3) and returns the cosine similarity as a single float in [-1, 1]. Useful for paraphrase detection, dedup, and cheap retrieval routing. Use it as an embedding similarity, semantic match, vector compare, or are-these-strings-similar check. | | employee-feedback-sentiment | (0.01 USDC/call) Scores internal survey comments and anonymized employee feedback by morale, workload, manager support, and retention risk. Same sentiment-analysis backend. Use it as an employee feedback sentiment API, engagement survey classifier, or workplace comment scorer. | | escalation-brief | (0.015 USDC/call) Turns a messy support thread into a short handoff brief the next person can act on in seconds instead of re-reading the whole conversation. Send the thread (a string or an array of messages) plus optional context, and get back an escalation summary: issue, impact, steps_tried, ask, urgency, and a suggested_owner when the thread implies one. Built for escalation summary, ticket brief, and support handoff workflows; every field is grounded strictly on the thread text, no invented steps or names. Redact customer or employee PII from thread text before sending it; this is a read-only summarization pass. Use it as an escalation summary API, ticket brief generator, support handoff tool, or incident summary endpoint for on-call and tier-2 handoffs. | | extract | (0.02 USDC/call) Pull structured entities out of raw text instead of hand-parsing it yourself. Send text and get back people, organizations, locations, dates, emails, urls, phone_numbers, and monetary amounts, each with a verbatim mention so you can locate it back in the source, plus a total_entities count. Amounts come with a parsed numeric value and currency code when the model can infer them, and every category returns an empty array rather than omitting the field when nothing matches. Use it as a named entity recognition API, NER extractor, or information extraction tool for contracts, emails, articles, or any text where you need names, dates, and numbers pulled out cleanly. | | extract-entities | (0.005 USDC/call) Pulls structured entities out of free text: people, organizations, locations, dates, email addresses, URLs, phone numbers, and monetary amounts, each returned verbatim as it appears in the source so you can locate it again. Send text and get back an entities object grouped by type plus total_entities, powered by an LLM extraction pipeline with automatic fallback if the primary provider is unavailable. Use it as a named entity recognition (NER) API, entity extractor, or information extraction tool for parsing contracts, emails, invoices, and news text into structured fields. | | finance-report-translate | (0.002 USDC/call) Translates finance reports: market commentary, investor letters, earnings notes, and finance summaries, preserving tickers and numbers. Same backend as translate-text. Use it as a finance report translation API, investor update localization tool, or financial memo translator. | | game-localization-translate | (0.002 USDC/call) Translates game content: dialogue, item descriptions, UI strings, and patch notes, preserving tokens. Same backend as translate-text. Use it as a game localization translation API, quest text translator, or in-game copy localization tool. | | hash-string | (0.005 USDC/call) Computes cryptographic hash digests of any input string across five algorithms: sha1, sha256 (default), sha384, sha512, and md5. Returns hex, base64, and base64url encodings of each digest. SubtleCrypto powers the SHA family; MD5 is computed by a small in-process implementation (the web crypto spec doesn't expose MD5, but it's still useful for cache keys and content fingerprinting where collision-resistance isn't required). Use it as a string hasher, multi-algorithm digest, cache-key generator, or content fingerprinter for SHA-256 / SHA-1 / SHA-384 / SHA-512 / MD5. | | hotel-review-sentiment | (0.01 USDC/call) Classifies hotel, short-term rental, and travel experience reviews by cleanliness, location, staff, amenities, and value. Same sentiment-analysis backend. Use it as a hotel review sentiment API, travel review classifier, or guest feedback scorer. | | humanize | (0.01 USDC/call) Rewrite AI-sounding text so it reads like a person wrote it. Send text with an optional tone of neutral, casual, professional, or academic and a preserve_length flag, and get back a rewritten text string that strips formulaic openers, bloated transitions, hedging phrases, corporate filler words, and overused em dashes while keeping the original meaning and facts intact. Use it as an AI text humanizer, GPT-detector evasion tool, or LLM-tell remover when AI-generated drafts need to pass as natural writing before publishing, emailing, or submitting. | | image-describe | (0.02 USDC/call) Get a vision model's read on an image: a description, alt text, extracted text, tags, or a caption. Send an image_url and a mode (describe, alt_text, ocr, tags, or caption), or override with a custom prompt, and it returns the generated text along with the mode used. OCR mode pulls out any visible text verbatim. Use it as an image captioning API, alt-text generator, image OCR tool, or vision-based tagging endpoint. | | image-description | (0.02 USDC/call) Takes a public image URL and returns an AI vision description, alt text, OCR text, tags, or caption depending on mode. Use it as an image description API, AI image captioner, or image-to-text endpoint. | | immigration-document-translate | (0.002 USDC/call) Translates immigration form excerpts, appointment instructions, evidence checklists, and application notes, preserving names, dates, IDs, and numbering. Not legal advice. Same translate-text backend. Use it as an immigration document translation API, visa form localization tool, or official document translator. | | invoice-translate | (0.002 USDC/call) Translates invoice notes, payment terms, receipt line descriptions, and billing messages, preserving amounts, invoice IDs, and dates. Same translate-text backend. Use it as an invoice translation API, receipt localization tool, or billing document translator. | | job-post-translate | (0.002 USDC/call) Translates job descriptions, role requirements, benefits, and candidate emails, preserving acronyms and compensation numbers. Same translate-text backend. Use it as a job post translation API, recruiting copy localization tool, or careers page translator. | | json-schema-validate | (0.003 USDC/call) Validates any JSON document against any JSON Schema, draft-07 or 2020-12. Returns valid boolean, total error count, and per-error instance_path + schema_path + keyword + message. Powered by @cfworker/json-schema (MIT), a spec-walker validator with built-in format keyword support (email / uri / date / uuid / etc.). Check whether an LLM's structured output matches the spec you asked for, validate an API response against your documented schema, or add a verification gate to an agent pipeline. 200KB schema cap + 500KB data cap keep CPU bounded. Use it as a JSON Schema validator, JSON validator, structured-output verifier, or contract checker. | | logo-prompt-generator | (0.01 USDC/call) Generates a ready-to-use logo prompt for a new brand. Same backend as brand-launch-brief; returns a compact logo_prompt plus name score, positioning, taglines, messaging pillars, and launch checklist for new brand workflows. Use it as a logo prompt generator, brand logo prompt source, or company logo creative brief. | | market-news-sentiment | (0.01 USDC/call) Market news sentiment analysis classifies supplied financial news headlines or market commentary as positive, negative, mixed, or neutral. Send text and optional aspects such as guidance or revenue; receive overall sentiment, a signed score, confidence, emotion probabilities, and per-aspect results. It analyzes caller-provided text and does not fetch news or provide investment advice. Use it as a market news sentiment API, financial news sentiment classifier, economic news tone scorer, or finance headline analysis tool. | | marketing-copy-translate | (0.002 USDC/call) Translates landing pages, ads, lifecycle emails, and social copy while preserving brand names and URLs. Same backend as translate-text. Use it as a marketing copy translation API, campaign localization tool, or ad copy translator. | | marketplace-listing-translate | (0.002 USDC/call) Translates marketplace listings: titles, bullet points, sizing notes, shipping terms, and product descriptions, preserving SKUs and units. Same translate-text backend. Use it as a marketplace listing translation API, ecommerce listing localization tool, or seller catalog translator. | | marketplace-notice-explain | (0.02 USDC/call) Explains marketplace and payment-platform notices for sellers. Parses pasted notices from Amazon, Etsy, eBay, Walmart, Shopify, Stripe, or PayPal and identifies likely notice type, deadlines, evidence checklist, and a response-packet outline. Explanation only: does not submit appeals, promise reinstatement, or bypass platform rules. Use it as a marketplace notice explainer or seller account warning triage. | | medical-intake-translate | (0.002 USDC/call) Translates patient intake notes, appointment messages, and care-navigation copy, preserving dates, names, units, and medication strings. Not medical advice. Same translate-text backend. Use it as a medical intake translation API, clinic form localization tool, or patient message translator. | | moderate-content | (0.02 USDC/call) Moderates content for safety, scoring harassment, hate_speech, violence, sexual_content, self_harm, spam, phishing, doxing, illegal_activity, plus custom categories, and returning allow/review/block. Morpheus primary, Venice fallback. Use it as a content moderation API, safety classifier, or OpenAI-style toxicity API. | | new-company-brief | (0.01 USDC/call) Builds a launch brief for a new company: positioning, taglines, messaging pillars, logo prompt, launch checklist, and channels. Same backend as brand-launch-brief. Use it as a new company brief API, business launch brief, or company positioning kit. | | paper-to-flashcards | (0.015 USDC/call) Turns a paper abstract or excerpt into study flashcards for spaced repetition. Send the text and an optional card count and get back question/answer/difficulty flashcards grounded only in what the text says, so you can drop them straight into a review deck without fact-checking against outside sources. This is paper to flashcards generation, not a summary of research you haven't given it. Use it as a paper to flashcards API, study cards generator, spaced repetition card builder, or learn-from-paper tool for researchers and students turning reading into review material. | | patient-feedback-sentiment | (0.01 USDC/call) Classifies patient comments by appointment access, staff, clarity, wait time, and trust. Classification only, not medical advice. Same sentiment-analysis backend. Use it as a patient feedback sentiment API, clinic review classifier, or healthcare experience scorer. | | pii-redact | (0.005 USDC/call) Redacts PII from text: structural PII (email, phone, credit card, SSN, IBAN, IPv4/v6, URL) and residual PII (full names, street addresses, dates of birth). Returns the redacted text and a list of every (type, value, masked_with) triple. Use it as a PII redactor, GDPR-safe text masker, or privacy scrubber for emails, phones, SSNs, IBANs, credit cards, and IPs. | | pr-description-from-diff | (0.01 USDC/call) Writes a review-ready pull request description from any unified git diff (up to 60k chars). Returns a structured PR title (<= 70 chars, imperative, conventional-commits-flavored), 2-5 summary bullets, a 3-7 item actionable test plan rendered as a Markdown checklist, plus a breaking_change boolean and migration note when applicable, and a ready-to-paste Markdown body with ## Summary, ## Test plan, and optional ## Breaking change sections. Pairs with commit-message-from-diff at a wider scope: commits get the one-line subject + body, PRs get the full review-ready narrative. Use it as an AI PR description generator, PR body from unified diff, GitHub PR body writer, GitLab MR description tool, or generate-pr-from-git-diff. | | product-localization-translate | (0.002 USDC/call) Translates product descriptions, SKUs, feature bullets, and launch copy while preserving Markdown, code, URLs, and product names. Same backend as translate-text, tuned for discoverability around ecommerce content. Use it as a product localization translation API, ecommerce copy translator, or app-store listing localization tool. | | prompt-compress | (0.005 USDC/call) Compresses a long prompt down to a target ratio of its original length while preserving every instruction, constraint, and example's intent. Drops filler words, redundant repetition, and ceremonial politeness. Powered by Venice mistral-small-3-2-24b. Use it as a prompt compressor, context shrinker, prompt distiller, or cost-cutter for long system prompts. | | reading-plan-generate | (0.015 USDC/call) Builds a sequenced reading plan for a topic so a learner or agent knows what to study first, next, and last. Send a topic plus an optional level and hour budget and get back a staged learning path: stage titles, goals, resource types to seek out, and estimated hours per stage. This is a reading plan generator and study plan generator, not a professionally curated or accredited curriculum, and it names resource types rather than fabricating specific unverifiable book titles or URLs. Use it as a reading plan API, learning path generator, study plan generator, or topic curriculum tool when onboarding into a new subject or budgeting study time. | | real-estate-lead-sentiment | (0.01 USDC/call) Scores real estate inquiries, showing feedback, and buyer/seller messages by urgency, objections, price sensitivity, and interest. Same sentiment-analysis backend. Use it as a real estate lead sentiment API, buyer message intent classifier, or property inquiry sentiment scorer. | | real-estate-listing-translate | (0.002 USDC/call) Translates real estate listings: property descriptions, amenities, neighborhood blurbs, and buyer messages. Same backend as translate-text. Use it as a real estate listing translation API, property description localization tool, or MLS copy translator. | | regex-from-prompt | (0.01 USDC/call) Builds a regular expression from a plain-English description, targeting PCRE, JavaScript, Python, Go, or RE2. Returns pattern, flags, explanation, and 3-6 test examples, and live-runs the JS regex on sample_text. Use it as a regex generator, NL to regex converter, or pattern builder. | | regex-test | (0.005 USDC/call) Tests a JavaScript regex against sample inputs and shows exactly what matched. Takes a pattern + flags and an array of test inputs, then returns a per-input matched / not-matched verdict, every match with byte index, all numbered capture groups, and named groups (?...). Catches catastrophic-backtrack-shaped patterns up front (nested unbounded quantifiers) and rejects them with a clear error. Companion to regex-from-prompt: that one generates a pattern from natural language, this one tells you whether your pattern actually matches what you think it matches. Use it as a regex tester, pattern matcher, regex playground, verify-a-pattern check, or match-and-capture extractor. | | restaurant-review-sentiment | (0.01 USDC/call) Classifies restaurant guest reviews by food, service, ambience, price, and return-intent aspects. Same sentiment-analysis backend. Use it as a restaurant review sentiment API, hospitality feedback classifier, or food-service review scorer. | | resume-scorer | (0.02 USDC/call) AI resume scorer / ATS keyword analyzer. Scores resume vs. job description (0-100 fit), with calibrated subscores: keyword match, experience, skills, formatting, impact. Ranked improvement suggestions. | | retrieval-rerank | (0.005 USDC/call) Reranks retrieval results: given a query and up to 30 candidate documents, scores each 0-100 for query relevance using Venice qwen3-5-35b-a3b in JSON-mode and returns sorted top-k. Useful as a second-stage reranker on top of cheap vector retrieval. Use it as a RAG reranker, document scoring endpoint, top-k filter, or cross-encoder substitute. | | review-sentiment-analysis | (0.01 USDC/call) Scores product reviews with overall sentiment, emotion probabilities, confidence, and optional aspect scores. Same backend as sentiment-analysis. Use it as a review sentiment analysis API, product review sentiment scorer, or app-store review classifier. | | rewrite-tone | (0.02 USDC/call) Rewrites text in a different tone or writing style. 12 tones: formal, casual, friendly, confident, empathetic, concise, playful, persuasive, apologetic, technical, simple, enthusiastic, with audience and length controls. Use it as a tone rewriter, paraphraser, or writing style changer. | | rfp-requirements-extract | (0.03 USDC/call) Extracts structured requirements from pasted RFP, grant, or local solicitation text: requirements, deadlines, attachment checklist, bid/no-bid flags, and missing organization-profile fields. File-in, JSON-out extraction only: no portal crawling, no bid submission, no award guarantee. Use it as an RFP requirements extractor, grant compliance matrix starter, solicitation parser, or bid checklist API. | | saas-ui-translate | (0.002 USDC/call) Translates SaaS UI copy: button labels, settings copy, onboarding text, and interface microcopy, preserving placeholders. Same backend as translate-text. Use it as a SaaS UI translation API, product interface localization tool, or app string translator. | | sales-call-sentiment | (0.01 USDC/call) Scores sales call snippets for objections, buying signals, urgency, and next-step confidence. Same backend as sentiment-analysis. Use it as a sales call sentiment API, prospect tone classifier, or transcript emotion scorer. | | semantic-chunk | (0.005 USDC/call) Splits long text into chunks for RAG pipelines, with three modes: 'fixed' (hard char-count windows with overlap), 'sentence' (greedy pack of sentences up to chunk_size), 'paragraph' (split on blank lines, never pack across paragraphs). Returns each chunk's text, start/end character offsets, and char count. Use it as a semantic chunker, text splitter, RAG chunker, or sentence + paragraph aware chunking-with-overlap tool. | | sentiment | (0.01 USDC/call) Score how positive, negative, or mixed a piece of text reads, down to the emotion level. Send text with an optional aspects array (up to 15 specific things to evaluate separately, like "service" or "price"), and get back overall_sentiment, an overall_score from -1 to +1, a confidence value, a short summary, and a per-emotion breakdown covering joy, anger, sadness, fear, surprise, and disgust. When aspects are supplied, each one gets its own sentiment, score, and supporting quote. Use it as a sentiment analysis API, emotion detection tool, or review and support-ticket triage system for gauging customer feedback or social mentions at scale. | | sentiment-analysis | (0.01 USDC/call) Analyzes sentiment in arbitrary text, returning overall sentiment, a score (-1 to +1), per-emotion labels (joy/anger/sadness/fear/surprise/disgust), and optional aspect-based scoring. A text-only analysis endpoint for product reviews, tickets, comments, and support logs. Use it as a sentiment analyzer, text sentiment classifier, emotion classifier, or aspect-based sentiment API. | | sla-extract | (0.015 USDC/call) Pulls the actual service-level terms out of a contract, vendor agreement, or support policy so you don't have to hunt through pages of text for them. Send the contract text (or a support policy excerpt) and get back structured SLA extraction: response_times, resolution_times, uptime_pct, penalties, escalation_tiers, and exclusions. Built for service level terms, uptime commitment, and escalation tiers lookups; this is text extraction only, not legal advice, and it doesn't judge enforceability. Redact customer or employee PII from the source text before sending it; the endpoint is read-only and reports only what the text actually states. Use it as an SLA extraction API, contract term parser, uptime commitment checker, or service-level terms lookup for vendor and support agreements. | | slugify | (0.005 USDC/call) Turns any string into a URL- and identifier-safe slug. Four modes: kebab (lower-kebab-case, default), snake (lower_snake_case), dot (lower.dot.case), or preserve_case (keep original casing). Unicode-aware: strips accents (José to jose), normalizes symbols to words for common cases (& to and, % to percent, @ to at, # to hash, + to plus), collapses adjacent separators, and strips leading/trailing separators. Optional max_length with word-boundary-aware truncation. Cheapest endpoint in the catalog at $0.001. Use it as a URL slug generator, slugifier, canonical-identifier maker, safe-string converter, SEO slug builder, filename slug, or cache-key normalizer. | | social-comment-sentiment | (0.01 USDC/call) Classifies social comments, replies, and threads by sentiment, emotion, confidence, and optional aspects. Same backend as sentiment-analysis. Use it as a social comment sentiment API, post reply emotion classifier, or community feedback scorer. | | social-sentiment | (0.03 USDC/call) Measures social sentiment on X for brands, tokens, topics, launches, and market narratives. Searches recent public posts and returns aggregate sentiment, tweet-level labels, distribution, average score, and public engagement metrics. Read-only X API v2 recent search; requires X_BEARER_TOKEN. Use it as a Twitter sentiment API or X sentiment search. | | sql-from-prompt | (0.02 USDC/call) Turns natural language into SQL for Postgres, MySQL, SQLite, BigQuery, Snowflake, MSSQL, DuckDB, or ANSI. Optional schema input; returns SQL plus explanation, tables_referenced, is_destructive, and warnings. Use it as a text to SQL converter, NL to SQL tool, or AI SQL generator. | | startup-launch-brief | (0.01 USDC/call) Generates a startup launch brief with positioning, taglines, and a launch checklist. Same backend as brand-launch-brief, with exact startup-launch wording for discovery. Returns structured positioning, taglines, messaging pillars, logo prompt, checklist, and channels. Use it as a startup launch brief generator, new company positioning kit, or founder brand brief. | | startup-slogan-generator | (0.005 USDC/call) Generates startup slogans and taglines for new-business launch copy, X bios, landing pages, and pitch decks. Same backend as brand-tagline-generate. Use it as a company slogan API or product tagline generator. | | structured-extract | (0.01 USDC/call) Extracts structured JSON from free-form text, conforming to a user-supplied JSON Schema. Powered by Venice zai-org-glm-4.7 (function-calling-default model) with response_format json_object. Returns the extracted object plus a 'validates' flag against the user's schema (basic top-level type and required-key check). Use it as a JSON-from-text extractor, schema-guided extraction, key-value pull, or form-filler. | | summarize | (0.01 USDC/call) Condense long text into a tighter summary in the shape you actually need. Send text (up to 30,000 characters) and a style of tldr, bullets, paragraph, or executive, with an optional max_words target, and get back a summary string sized and structured to match. TLDR mode gives a one- or two-sentence takeaway, bullets gives 3-7 punchy points, paragraph gives one cohesive block, and executive gives a headline, key points, and a recommendation. Use it as a text summarizer, article condenser, or executive-summary generator for reports, articles, transcripts, or any long-form content you need to skim fast. | | summarize-text | (0.01 USDC/call) Condense long text into a shorter summary. Send text (up to 30,000 characters) with an optional style, choosing from tldr, bullets, paragraph, or executive, plus an optional max_words target, and it returns a summary sized and structured to match, along with input and output character counts. Executive style adds a headline, key points, and a recommendation line. Use it as a text summarizer, article condenser, or executive summary generator for reports and long documents. | | support-sentiment-analysis | (0.01 USDC/call) Scores support tickets, chats, and complaints for sentiment and emotion, for helpdesk and CX agents. Same backend as sentiment-analysis. Use it as a support sentiment analysis API, customer ticket emotion classifier, or support escalation signal. | | support-ticket-translate | (0.002 USDC/call) Translates support tickets, replies, and customer feedback while preserving formatting and technical identifiers. Same backend as translate-text, exposed for support workflows. Use it as a support ticket translation API, customer support localization tool, or helpdesk message translator. | | survey-sentiment-analysis | (0.01 USDC/call) Scores open-ended survey responses by theme or aspect, for research agents. Same backend as sentiment-analysis. Use it as a survey sentiment analysis API, NPS verbatim classifier, or feedback sentiment scoring tool. | | syllabus-parse | (0.015 USDC/call) Turns a pasted course syllabus into a structured week-by-week outline an agent can schedule against: course name, and per week the topics, readings, assignments, and due dates. This is syllabus parse and course outline extract from the text you send: it reports what the syllabus states, including the reading schedule and assignment dates, and never fills in a week's contents that isn't in the source. Makes no claim about credit or accreditation. Use it as a syllabus parse API, course outline extraction tool, reading schedule parser, or assignment due-date extractor for study planning and calendar sync. | | tagline-generator | (0.005 USDC/call) Generates short tagline options from a name, concept, audience, and tone. Same backend as brand-tagline-generate, exposed under a broad buyer slug. Use it as a tagline generator, startup slogan generator, or brand positioning line API. | | technical-docs-translate | (0.002 USDC/call) Translates developer documentation with code-aware handling that preserves Markdown, code blocks, API identifiers, URLs, and file paths. Same backend as translate-text. Use it as a technical documentation translation API, developer docs localization tool, or README translator. | | text-classify | (0.005 USDC/call) Text classifier. Zero-shot classification — caller supplies 2-25 labels and the model picks the best one (or all matching labels in multi-label mode) with calibrated confidence scores and a per-label reason. Same backend as classify-text under a more search-friendly slug. Mistral powered. | | text-embedding | (0.005 USDC/call) Embeds 1 to 100 strings into semantic vectors via Venice. Tier shorthand: 'default' → gemini-embedding-2-preview (newest, recommended), 'fast' → text-embedding-bge-m3, 'openai-compat' → text-embedding-3-small. You can also pass a full Venice embedding model name. Returns a list of vectors aligned with input order. Use it for text embedding, vector embedding, Venice embeddings, Gemini embeddings, or BGE-M3. | | text-normalize | (0.005 USDC/call) Text normalize. Unicode NFC / NFD / NFKC / NFKD normalization plus per-codepoint script classification (Latin / Cyrillic / Greek / Hebrew / Arabic / CJK / etc.), homoglyph detection (the Cyrillic 'а' that looks like Latin 'a'), and invisible / RTL / BOM detection. Same backend as unicode-normalize. | | thesaurus | (0.005 USDC/call) Returns synonyms, antonyms, and related words for any input word, configurable across five modes: 'synonyms' (rel_syn), 'antonyms' (rel_ant), 'sounds_like' (homophones / rhymes), 'similar_meaning' (semantically near), or 'related' (commonly co-occurring triggers). Wraps the public Datamuse API, no auth, commercial-OK. Use it as a thesaurus, rhyme finder, paraphrasing aid, or query expansion tool. | | ticket-cluster | (0.02 USDC/call) Groups a batch of raw support tickets into themed clusters so you can see what's actually driving volume without reading every ticket by hand. Send an array of ticket texts and get back labeled clusters, each with a ticket count, the original ticket_indexes it covers, and a representative sample, built for ticket clustering, support theme grouping, and ticket triage at queue scale. Grounded strictly on the ticket text you send, up to 200 tickets per call. This is read-only analysis; redact customer PII from ticket text before sending it. Use it as a ticket clustering API, support theme grouping tool, ticket triage assistant, or issue clustering endpoint for high-volume support queues. | | token-count | (0.005 USDC/call) Estimates LLM token counts entirely locally, with no external lookups. Heuristic estimator targeting cl100k_base / o200k_base (GPT-4o) with calibrated multipliers for Claude and Gemini; the input model name is matched against an internal table of supported tokenizers. Accuracy ±5% versus tiktoken on typical English. Use it as a tokenizer estimate, GPT-4 token count, Claude token count, Gemini token count, or context-window pre-flight. | | tool-card-generate | (0.005 USDC/call) Generates an agent tool card from a tool name + plain-English description (and optional parameter hints): a strict OpenAI/A2A-compatible spec of { name, description, parameters: }. Powered by Venice zai-org-glm-4.7 in json_object mode. Useful for retrofitting LLM-callable tool descriptions onto existing API endpoints. Use it as an OpenAI function-calling spec generator, A2A tool-card builder, or agent tool description writer. | | translate | (0.002 USDC/call) Fast machine translation API for agents. Send text and a target_language, with optional source_language and a casual, formal, or neutral register. It returns translated_text plus the detected source language and preserves Markdown, code blocks, URLs, and proper nouns. Uses a latency-tuned translation model with an automatic fallback across 100+ languages. Use it for chat, documentation, support, and localization workflows. | | translate-text | (0.002 USDC/call) Translate agent messages, support replies, and documents without breaking Markdown, code blocks, URLs, or proper nouns. Send text and target_language, then get translated_text plus detected_source_language; source_language and formality are optional. It auto-detects the source and uses a latency-tuned fallback for 100+ languages. Use it as a translation API, document translation service, support reply translator, or localization API for recurring agent workflows. | | travel-listing-translate | (0.002 USDC/call) Translates travel and hospitality copy: lodging descriptions, itinerary copy, amenities, and guest messages. Same backend as translate-text. Use it as a travel listing translation API, hotel and tour localization tool, or hospitality copy translator. | | tweet-sentiment | (0.03 USDC/call) Analyzes the sentiment of recent tweets matching a query. Fetches recent public X posts, excludes retweets, and returns per-tweet labels plus aggregate sentiment distribution and average score. Use it as a tweet sentiment analyzer, Twitter recent-search sentiment tool, or X post sentiment API. | | twitter-sentiment | (0.03 USDC/call) Searches recent public X posts for a query and scores each tweet's sentiment, returning tweet-level scores plus aggregate positive/neutral/negative distribution and average score. Excludes retweets by default with an optional language filter. Requires X_BEARER_TOKEN; read-only, no posting, replying, liking, following, or account mutation. Use it as a Twitter sentiment API, X sentiment API, social sentiment monitor, tweet sentiment analysis, brand sentiment search, or crypto sentiment tracker. | | twitter-sentiment-api | (0.03 USDC/call) Analyzes Twitter/X sentiment by searching recent posts and scoring sentiment, polarity, volume, and sample post context when X API credentials are configured. Same X-backed handler as social-sentiment. Use it as a Twitter sentiment API, X sentiment analysis, or social sentiment endpoint. | | type-inference-from-json | (0.005 USDC/call) Infers type definitions from a JSON sample, converting JSON to TypeScript, Zod, or JSON Schema. Paste a sample (or array of samples); returns a generated type definition in your chosen format. Merges across array elements / object samples: properties present in some but not all become optional, mixed primitive types become unions. Auto-detects string formats (date-time / uuid / uri / email) for richer outputs. Ideal inside agent code-gen loops that need to consume an unfamiliar API response. Use it as a JSON shape inferer or quicktype-style type generator. | | unicode-normalize | (0.005 USDC/call) Normalizes Unicode text and flags lookalike or hidden characters used in spoofing and phishing. Normalizes to NFC (default), NFD, NFKC, or NFKD, classifies every codepoint by script (Latin / Cyrillic / Greek / Hebrew / Arabic / CJK / Hangul / etc.), flags Cyrillic / Greek / Latin Extended homoglyphs (the Cyrillic 'а' that looks like Latin 'a', etc.) with their position, codepoint, and the ASCII char they impersonate, and surfaces hidden / formatting characters like zero-width spaces, RTL overrides, and BOMs. Use it for homoglyph detection, IDN spoof checks, invisible-character and zero-width scans, and phishing detection. | | voice-of-customer-sentiment | (0.01 USDC/call) Processes survey verbatims, support snippets, reviews, and interview notes by theme, sentiment, emotion, and urgency. Same sentiment-analysis backend. Use it as a voice of customer sentiment API, VOC feedback classifier, or customer insight scoring tool. | | x-sentiment | (0.03 USDC/call) Scores sentiment across recent X posts for any query, returning positive, neutral, and negative labels with result counts and aggregate distribution. Runs X API v2 recent search and scores public posts locally. Use it as a Twitter sentiment monitor or social listening API. |

How it works

1. Agent calls a tool (e.g. ai-to-human-text).
2. MCP server POSTs to https://x402.agentutility.ai/ai-to-human-text.
3. The endpoint responds HTTP 402 with payment instructions.
4. The MCP server signs an EIP-3009 USDC transfer authorization with X402_PRIVATE_KEY and retries.
5. CDP facilitator settles on Base.
6. The endpoint returns the actual response.

The agent never sees the payment flow — it just gets the result.

Links

• Cluster overview: https://agentutility.ai/wordmint/
• All MCP packages: https://mcp.agentutility.ai/
• Source: https://github.com/rooz21/x402/tree/main/packages/mcp-wordmint

Version: 0.17.3 · License: MIT