Fyrmo
Server Details
Small businesses in Costa Rica, Panama, Guatemala and Dominican Republic: WhatsApp, phone, email
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-11-25
- URL
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 12 tools
Most tools have clearly distinct purposes: search vs. segment sizing vs. matching vs. similar-business discovery vs. quote/build pipeline. The main overlap is contact reveal across build_list, reveal_contacts, and get_business, but the descriptions make the intended boundaries reasonably clear.
Almost all tools use snake_case and a predictable verb_object pattern (search_businesses, quote_list, build_list, reveal_contacts, report_outcome). The only notable deviation is the noun-phrase my_contacts, which is still understandable.
Twelve tools is well-scoped for a business-search, list-building, and contact-reveal workflow. Each tool has a clear role, with no obviously redundant additions or missing operational pieces.
The surface covers discovery (search, categories, segments), pricing and purchasing lists, targeted reveal, record lookup, CRM matching, similar-business expansion, credit purchase, contact history, and outcome reporting. No major lifecycle gaps are apparent for the stated domain.
Available Tools
12 toolsbuild_listBuild a contact listADestructiveIdempotentInspect
Buys the list priced with quote_list: same filters and count, plus max_credits, the most the person agreed to spend. Reveals the businesses in order of record quality, never spends more than max_credits (businesses already revealed come free) and returns one row per business (WhatsApp with its wa.me link, mobile, phone, email, website, place, when the contact was last seen) plus a receipt: delivered, charged, free, how many have WhatsApp, credits per WhatsApp and credits left. Only businesses with at least one contact are listed; to pay only for WhatsApp, set channel: 'whatsapp'.
Inputs: count (required), max_credits (required), name, place, +15 more.
buy_creditsBuy creditsARead-onlyIdempotentInspect
Credit packs (100 = US$29, 500 = US$119, 2,000 = US$399, 10,000 = US$1,490) with their Stripe payment link. The link carries the client number, so credits load on their own once the payment completes. Give the link to the person as is; Fyrmo never asks for card details here.
Inputs: no named parameters.
find_similarFind similar businessesARead-onlyIdempotentInspect
Businesses like a given one: same industry, categories in common, nearby (same canton, then province) and the same signals (website, Instagram/TikTok, how they book, independent). Use it to grow a list from a business that worked. No credits; counts toward the daily search results; contacts are not shown.
Inputs: code (required), limit.
get_businessGet businessARead-onlyIdempotentInspect
The full record of one business (code) or up to 10 (codes): location, description, opening hours (OpenStreetMap format), its website in detail (what it is built with, last update, https, mobile version, online store, booking system, travel sites, delivery apps, payments, chat, analytics and ad pixels), social profiles, attributes with their source (services, prices, how it takes bookings) and its contacts. No credits, up to 300 a day (20 until the account buys credits); revealed ones don't count. Contacts stay masked and social profiles hidden (value null, hidden true) until revealed.
Inputs: code, codes.
list_categoriesList industries and categoriesARead-onlyIdempotentInspect
Every industry and category slug Fyrmo uses, with its English and Spanish name and how many businesses it has (Costa Rica and Panama together). Use the slugs as industry and category in the other tools. No credits, works without a key.
Inputs: industry.
match_listMatch a list of businessesARead-onlyIdempotentInspect
Finds which businesses of a list the person already has (a CRM export, a spreadsheet) are in Fyrmo. Each row matches by phone, email or own website, or by exact name plus canton or district; a phone or website shared by several businesses (a chain, a directory) only counts when the name tells them apart. Returns, per row (0-based), the business code and what it matched by, or why not (no_data, not_found, ambiguous, daily_limit), plus the matched businesses without their contacts. No credits; counts toward the daily search results; reveal the ones they want with reveal_contacts. Up to 200 rows per call. The list is not stored.
Inputs: rows (required).
my_contactsMy contactsBRead-onlyIdempotentInspect
The businesses you revealed with their contacts and wa.me links, the last outcome you reported and when the reveal expires (after that the contacts come back masked until revealed again), plus your balance.
Inputs: limit, offset, status.
quote_listQuote a contact listARead-onlyIdempotentInspect
Prices a list before buying it. Takes the same filters as search_businesses plus how many businesses you want (up to 200) and says how many match, how many of the first count with contacts you already revealed (free again) and how many credits the rest would cost, with a 10-row sample. Spends no credits. Show the person the price, then call build_list with the same filters.
Inputs: count (required), name, place, social, +14 more.
report_outcomeReport outcomeAInspect
Records how the contact with a revealed business went. wrong_number and closed give the credit back (within 14 days, up to 10% of what was revealed that month) and send the business back to research. Use note 'do not contact' when a business asks not to be written to again.
Inputs: code (required), outcome (required), note, contact.
reveal_contactsReveal contactsADestructiveIdempotentInspect
Returns the WhatsApp, mobile, phone and email of up to 50 businesses, each with its confidence, sources and last-seen date, and a wa.me link for every WhatsApp or mobile so the person can write one to one (their social profiles then show in get_business). Costs 1 credit per new business; businesses already revealed (12 months) and businesses without contacts are not charged. Limit: 500 per day.
Inputs: codes (required).
search_businessesSearch businessesARead-onlyIdempotentInspect
Search businesses in Costa Rica (or Panama with country: 'PA', Guatemala with 'GT', the Dominican Republic with 'DO') by industry, category, place, name and contact channel. Costs no credits: says which contacts each business has without revealing them, sorted by record quality. Combine industry and place with website, social, booking, min_rooms, email_type, ad_pixel, online_store, built_with and outdated_site to find the right target (spas that book on WhatsApp, businesses on Instagram with no website, stores that use Gmail, Wix sites not updated in years). Each result has signals with the same words: website, social, booking, booking_systems, rooms, email_type, ad_pixels, online_store, built_with, site_year (last update), no_mobile_version, no_https and opening_hours (OpenStreetMap format). Returns total and next_offset (null on the last page). Up to 3,000 results a day (100 until the account buys credits) and the first 1,000 of each search. detail: 'summary' drops signals and categories to save tokens.
Inputs: name, limit, place, detail, +16 more.
size_segmentsSize segmentsARead-onlyIdempotentInspect
How many businesses each segment has and how many of them have WhatsApp, without listing them. No credits, and doesn't count toward the daily search results. Use it to choose who to target before searching: e.g. for someone who sells espresso machines, compare cafés, bakeries, hotels and restaurants in a place. A segment takes the same filters as search_businesses. Up to 6 segments per call; results come back in the same order.
Inputs: segments (required).
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
Costa Rica local business directory — verified businesses, jobs, events, specials. Read-only.
Sell on WhatsApp and get paid in LATAM: CRM, payment links, collections, CFDI invoices, store.
Official public data for Central America and the Dominican Republic, down to the municipality.
Official public data for Central America and the Dominican Republic, down to the municipality.
Related MCP Servers
- FlicenseNot gradedqualityCmaintenanceEnables AI assistants to query official public data for Central America and the Dominican Republic down to the municipality level, covering indicators, prices, procurement, legal sources, news, and research with cited results.-
- FlicenseNot gradedqualityBmaintenanceConnects and syncs e-commerce stores (Shopify, WooCommerce, Google Merchant Center) directly to WhatsApp Shop Manager to automate product catalogs and messaging.-
- AlicenseCqualityDmaintenance244 WhatsApp Business tools for AI agents: send messages, automate workflows, run marketing campaigns, and manage CRM. Supports Streamable HTTP and stdio transports.1002MIT
- AlicenseAqualityDmaintenanceMCP server for Karrito - the digital catalog builder for WhatsApp sellers in LATAM, enabling AI assistants to manage store operations like products, orders, discounts, reviews, shipping, and analytics.3020 npmMIT
Glama MCP Gateway
Add one secure layer between your agents and this server.