MCP App Store
by KeenEthics
Licto icon

Licto MCP Server for Claude and ChatGPT

By Licto

Overview

Licto is an analysis platform for Spanish public procurement: tenders (licitaciones), the bodies that buy (organismos) and the companies that win (adjudicatarias), from official sources of the State, the autonomous communities and local councils. With this connector, your AI assistant queries that data in plain language. What you can do: - Search tenders by text, region or province, CPV code, budget, deadline, procedure and contract type, and open the detail: award criteria, solvency requirements, lots, official documents (pliegos) and the award results of each lot (winner, amount, bids received, lowest and highest bid). - Find similar tenders and who won comparable contracts. - Look up contracting bodies and winning companies: tender volume, budgets, contracts won and amount awarded. - See a company's awards lot by lot, with date, buyer, amount, discount to the budget (baja) and bids received, or grouped by buyer with its first and last award. - See market statistics for Spain, a sector or a region, for all time or for a year, quarter or month: median budget, open tenders, average bids and the share with a single bid, and the most active companies and buyers. - Compute the affinity score between a tender and your company profile or, on paid plans, a profile you describe, such as a client's. It is a relevance score, not a recommendation to bid. Every result links to its page on Licto and to the data of its source: official notice, license and how to cite it. The deadlines that count are the ones in the official notice. The full list of sources is at licto.es/fuentes. All tools are read-only: the connector does not submit bids or change anything in your account. You need a Licto account, which is free. You sign in with OAuth and can disconnect anytime under Cuenta → Conectores IA (Account → AI connectors). The data and the Licto pages are in Spanish.

How to connect

  1. Claude: Open the Licto listing in the Claude connectors directory and choose Connect. Claude listing Licto exposes 14 tools here.
  2. ChatGPT: Open the Licto listing in the ChatGPT app directory and choose Connect. ChatGPT listing Licto exposes 14 tools here.

Tools

get_company

ChatGPT
Get one company's profile summary: wins, awarded volume, location. total_awarded_eur is the NET (excluding VAT) volume. Awards whose source publishes only a VAT-inclusive figure (some regional portals) are not in it: their VAT-inclusive euros come apart as vat_inclusive_only_awarded_eur; never add them to the net total. ute_participation (when the company is a member of UTEs) states those UTEs' awards at their full contract value, apart from wins; group lists the company's other legal entities (same entity under another legal form) with their own figures. Neither is added to wins or total_awarded_eur. include=['activity'] adds by_year (wins and awarded euros per year; dated awards only, so it sums to less than wins when some awards carry no date), by_month (last 24 months), the first and last award date, the top 10 buyers (top_authorities, of top_authorities_total), top_regions, top_cpv and by_contract_type. One row per awarded lot: get_company_awards.

get_company_awards

ChatGPT
List the awards of one company: ONE ROW PER AWARDED LOT (a tender won in three lots is three rows) with award_date, lot and lot_title, buyer (id, name, url), region, cpv, award_amount_eur (the company's own share, net of VAT), budget_eur, baja_pct (discount of the award to the budget it was bid against, only when it is a plain 0-95 % discount), received_bids, segment and the tender's url. Find the company id with search_companies. The rows are the lines get_company.wins counts: contratos menores are included (is_menor) unless exclude_menores, and awards the source gave no date (award_date_missing) are included unless a date bound is given (undated_rows counts them). award_amount_eur is absent for a framework agreement (framework_ceiling_eur: the ceiling is shared with the other winners), a placeholder or a figure published only with VAT (award_amount_with_vat_eur): see amount_note. group_by='buyer' gives one row per buyer instead, with wins (tenders), awarded_eur, first_award_date and last_award_date: who the company sells to and since/until when. Filters: awarded_after / awarded_before (the line's award date), cpv, comunidad, organization_id. Newest award first, or sort=amount_desc. Pages of up to 50: while the result has a next_cursor, pass it as cursor (same other arguments); total is the number of rows and stops counting at 10000, so narrow with dates past it. For one row per TENDER with its status use search_tenders(company_id=...).

get_glossary

ChatGPT
Return the Licto glossary: what every grade, index and competition metric MEANS (LOF/LSI indicators, afinidad, single-bid rate, status values, affinity factors, conventions). Call this when unsure how to interpret a Licto field.

get_market_stats

ChatGPT
Market KPIs for the global market, a sector or a comunidad, over the history Licto has processed (partial, not every published tender; per-source years at https://licto.es/fuentes#historico): all of it by default, or one period ('year' / 'quarter' / 'month' of tender PUBLICATION; period_start picks it, else the newest). A period row has period_start and is_final: false while the period is open, when its figures and top lists are partial. Tender volume (listable_count = tenders search can return; tender_count also includes contratos menores), open tenders now (open_count), budgets (prefer the medians), competition (avg bids, single-bid rate), top_contractors (company_id, name, wins = award lines won, awarded_eur; frameworks, personas físicas, hidden companies and public bodies excluded, so wins is not get_company's), top_authorities (organization_ids, tenders, budget_eur) and the procedure_type_dist / contract_type_dist counts. See get_glossary.

get_match_score

ChatGPT
Compute the user's affinity score (0-100) for a tender from their Licto company profile (sector, region, budget range, contract types) with the matched-factor reasons. A relevance score; not a recommendation to bid.

get_organization

ChatGPT
Get one organization's profile summary: tender volume, budgets, contact.

get_similar_tenders

ChatGPT
Find tenders similar to a given one — useful as references (including who won comparable contracts). Each row has its source, source_url and source_updated_at; the sources block gives the publisher, licence and attribution text per source, with provenance_notice.

get_tender

ChatGPT
Get the detailed summary of one tender: dates, procedure, CPV, lots; amounts (budget_eur = presupuesto base de licitación excluding IVA as the source publishes it, budget_with_vat_eur = the same with IVA when stated, estimated_total_eur = valor estimado del contrato excluding IVA); award criteria with weights (criteria), solvency requirements, guarantee and variants, the notices published (lifecycle_events) and contract modifications; award results (award_results: one row per result the source published, with the lot number and title, result_code and its label result — adjudicado, desierto, desistimiento, renuncia —, winner, amount, bids received, SME bids, lowest/highest bid, abnormally-low flag) and lots_without_result (the lots of a partly awarded tender with no result published: not awarded, or not published yet). An amount of 0 on a framework agreement or a placeholder amount is absent, with an amount_note. publication_date is absent when the source published none. Long lists are capped: lists_truncated gives the full count. Official text (title, description_official) and Licto AI text (title_ai, description_ai) come in separate fields; provenance names the source, licence and attribution to cite; synced_at is when Licto last saw the record in its source.

get_tender_documents

ChatGPT
List a tender's official documents (pliegos, annexes) with download URLs; the documents themselves are read from those links.

list_sectors

ChatGPT
List Licto's sector taxonomy with each sector's CPV divisions. Use to translate a plain-language activity ('road construction') into a CPV code or sector slug for precise filtering.

score_tender

ChatGPT
Score a tender (0-100) against an EXPLICITLY-GIVEN company profile (sectors, regions, contract types, budget range) instead of the connected user's stored profile. Use this to evaluate fit for a client or a hypothetical company — the company need not have a Licto account.

search_companies

ChatGPT
Search companies (contractors/empresas) that win Spanish public contracts. Returns key aggregates per company.

search_organizations

ChatGPT
Search contracting organizations (organismos públicos) that publish tenders. Returns key aggregates per organization.

search_tenders

ChatGPT
Search Spanish public procurement tenders (licitaciones) with full-text query and filters (place by comunidad or provincia, CPV, budget, dates, procedure and contract type, framework, EU threshold) and a choice of sort order (e.g. sort=budget_desc for the largest). Returns compact rows with licto.es links and, per tender, its source (source, the record's source_url and source_updated_at); the response's sources block, once per source, gives the publisher, licence and attribution text to cite, with provenance_notice. Amounts: budget_eur is the presupuesto base de licitación excluding IVA, as the source publishes it (a source that publishes only a VAT-inclusive figure gives that one); estimated_total_eur is the valor estimado del contrato (excluding IVA, with options and extensions) when the notice states it. min_budget/max_budget filter budget_eur. publication_date is absent when the source published none (Licto then holds only a stand-in, the award or formalization date, which is not when the tender was published); published_after leaves such tenders out, so with it total can count tenders that never reach the pages, and a page can be shorter than limit while next_cursor still leads to more. cpv: a 2-digit division, a 3-7 digit prefix or a full 8-digit code, each matching its descendants; several codes comma-separated (up to 10), any of them matches. query: plain words, every word must match (no OR, no phrases); run separate searches for alternatives. comunidad / provincia filter by the NUTS region recorded in the notice. Tenders recorded only as nationwide ('España', typical of national bodies) are listed under every comunidad; one whose notice records another region, or none, can be missed, so complement the filter with a query naming the place. An awarded tender's row adds winner, winner_company_id, award_amount_eur (the award of the whole tender, absent when the source states none; get_tender gives it per lot), received_bids and award_date; a tender with no award has none of them. To list the tenders a company won (one row per tender; one row per awarded lot is get_company_awards): find its id with search_companies, then call search_tenders(company_id=...); awarded_after / awarded_before bound the award date (only with company_id). In that mode the rows come from the award data, not the text index: query and sort=relevance are refused, every other filter applies (status, comunidad / provincia, cpv, budgets, deadline_*, published_after, organization_id, procedure_type, contract_type, is_framework, over_threshold, sort), winner is that company, award_amount_eur is what it was awarded (a multi-lot tender may have other winners: see get_tender) and total stops counting at 10000. Pages of up to 50: while the result has a next_cursor, pass it as cursor (same other arguments) for the next page; total is the estimated number of matches.

get_company

Claude

get_company_awards

Claude

get_glossary

Claude

get_market_stats

Claude

get_match_score

Claude

get_organization

Claude

get_similar_tenders

Claude

get_tender

Claude

get_tender_documents

Claude

list_sectors

Claude

score_tender

Claude

search_companies

Claude

search_organizations

Claude

search_tenders

Claude

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/licto"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/licto)