io.github.scraping-api-plans/web-data-plan-finder

Est-ce que io.github.scraping-api-plans/web-data-plan-finder est sûr ?

Au dernier audit MCP Hub, le serveur MCP « io.github.scraping-api-plans/web-data-plan-finder » est classé « Non encore testé ». Référencé dans l'annuaire. Seul le pentest comportemental fait foi, et il n'a pas encore été exécuté sur ce serveur. Le verdict est réévalué en continu ; consultez la fiche pour l'état à jour.

Méthode : MCP Hub analyse la description de chaque outil (couche 1) et, pour le verdict fort, exécute le serveur dans une micro-VM isolée avec leurres et réseau bloqué (couche 2). On publie des noms et des compteurs, jamais le contenu brut renvoyé par un outil.

Les 6 outils

• check_conditions≈ 11422 tokens

  Call this right before choosing or buying an API plan: it returns, for each plan, whether it meets every condition, the monthly cost, and the official sentence behind each verdict. Use this when: the user names specific plans (by name, id or pricing-page URL) and wants to know whether they meet the workload and conditions, or which plan of the same product is the cheapest that does. Do not use this when: the user has no plans in mind yet (use search). Call it right before you choose or buy a tool or plan; format "brief" returns a short answer. Check specific plans, given by plan id, plan name or pricing-page URL, against the workload and conditions; up to 20 plans in one call are judged against the same conditions. Returns met / not met / unknown per condition with the official source sentence, which plans and billing options each value applies to, and a purchase link. For each plan it also names the cheapest plan of the same product that meets every required condition, with its monthly cost, or null if none does. An ambiguous name or URL returns the candidate plans instead of a guess. If you already have candidate plans or providers (for example from a web search), pass them as providers (search) or offer_ids (check_conditions) so that only those are judged. Example: 2 plans by id against 100,000 emails a month with a dedicated IP.

  Paramètres, retour & exemple (58)

  Paramètres

  • plans · array · optionnel
    Plans given by provider and plan name (e.g. "ScrapingBee Freelance") or by pricing-page URL, for plans found elsewhere. A name or URL that fits several plans returns the candidates under resolution and no verdict.
  • detail · string · optionnel · summary | full
    summary returns one source sentence per fact; full returns every recorded fact and sentence.
  • format · string · optionnel · full | brief
    brief returns, per plan, the plan, provider, whether it meets all, the monthly cost and the requested conditions only, each with its verdict, a one-line reason, one source sentence and its source (about a quarter of the full size); the verdicts and costs are those of full. Call it right before choosing or buying a plan.
  • mcp_auth · array · optionnel
    Ways to authenticate to the provider's hosted MCP endpoint, any one of which will do: api_key, oauth. Met when the endpoint accepts at least one of them; implies a hosted endpoint.
  • needs_zdr · boolean · optionnel
    Zero data retention must be usable on this plan. A provider that offers it only on a custom enterprise agreement does not meet it on its listed plans.
  • offer_ids · array · optionnel
    Plan ids to check (from search or get_details).
  • preferred · object · optionnel
    Nice-to-have conditions; they do not filter, they break ties.
  • needs_soc2 · boolean|string · optionnel · True | False | II
    The provider must hold a SOC 2 report: true for Type I or Type II, "II" for Type II.
  • needs_batch · boolean · optionnel
    The plan must transcribe recorded audio (a file or a stored recording), for speech-to-text. A plan priced for live audio only is not met.
  • data_regions · array · optionnel
    Regions where data must be stored or processed, e.g. ["EU"]. Different from exit-IP countries.
  • needs_korean · boolean · optionnel
    The service must voice Korean text (text-to-speech) or transcribe Korean speech (speech-to-text).
  • billing_types · array · optionnel
    Acceptable billing: monthly_subscription (A plan with a fixed fee (or minimum commitment) billed every month.) annual_subscription (A plan paid for a year at a time.) pay_as_you_go (No plan fee; a price per unit of use, billed after use or drawn from a balance topped up with any amount.) prepaid_credits (Money paid before use: a credit pack, a balance topped up first, or a plan amount charged in advance for the period.) Free plans always qualify.
  • monthly_pages · object · optionnel
    Pages per month by type: plain_html (static pages), javascript (needs rendering), premium_proxy (blocked sites that need residential/premium proxies), javascript_premium_proxy (both).
  • needs_inbound · boolean · optionnel
    The service must receive email for the buyer and hand it on (inbound parsing, routes), for transactional email.
  • monthly_emails · number · optionnel
    Emails sent per month, for transactional email APIs (SendGrid, Mailgun, Postmark, Resend, Amazon SES, Brevo, Mailtrap, Loops). It is the whole workload: give it instead of monthly_pages, monthly_searches and monthly_browser_hours, not with them. Only email products can serve it; other products are out of scope.
  • search_engines · array · optionnel
    Search engines whose results the plan must return, each at the plan's search price: google, bing, yahoo, yandex, baidu, duckduckgo, naver, brave (brave is Brave's own index). monthly_searches is the total over all of them.
  • captcha_solving · boolean · optionnel
    The plan must solve CAPTCHAs for the browser sessions (the vendor does it, not the buyer own service).
  • min_concurrency · integer · optionnel
    Minimum number of concurrent requests (or browsers) the plan must allow.
  • needs_streaming · boolean · optionnel
    The service must return audio in real time as it is made (streaming or a WebSocket), for text-to-speech; for speech-to-text, it must transcribe live audio as it arrives. A speech-to-text price line is per mode, so a plan priced for recorded audio only is not met.
  • monthly_searches · integer · optionnel
    Search requests per month (one results page each).
  • needs_remote_mcp · boolean · optionnel
    The provider must host a remote MCP endpoint (a URL an MCP client connects to). An official server that can only be run on the buyer own machine does not meet it.
  • proxies_included · boolean · optionnel
    The plan must include proxy traffic of the vendor (some gigabytes in the plan fee). Proxy traffic sold only per gigabyte on top of the fee, or only on another plan, does not count.
  • search_countries · array · optionnel
    ISO 3166-1 alpha-2 codes of the countries whose search results the searches must return (the search engine's country setting, e.g. Google gl), e.g. ["GB","CA"]. Different from exit_ip_countries.
  • exit_ip_countries · array · optionnel
    ISO 3166-1 alpha-2 codes of countries the page requests must exit from, e.g. ["US","DE"]. For search results of a country use search_countries.
  • min_pages_per_doc · integer · optionnel
    Pages one document may have at the most the plan allows, for document parsing. Met when the stated limit is at least this many (or the pages say there is no page limit); a plan whose pages do not state a limit is unknown. The limit of the largest way of sending (batch or asynchronous) counts.
  • needs_diarization · boolean · optionnel
    The plan must label who spoke (speaker diarization), for speech-to-text. Where the vendor charges extra for it, the extra price is added to the monthly cost; where the pages state no price or no inclusion the plan is unknown.
  • needs_no_training · boolean · optionnel
    An official page must say the vendor does not use the buyer's text, audio or documents to train its models without the buyer opting in. A use that is on by default with an opt-out to ask for does not count.
  • session_recording · boolean · optionnel
    The plan must record browser sessions (video or replay) that the buyer can watch afterwards.
  • monthly_characters · number · optionnel
    Characters of text voiced per month, for text-to-speech APIs (ElevenLabs, Cartesia, Deepgram Aura, OpenAI, Google Cloud, Amazon Polly, Azure, Gradium, Rime, Inworld, MiniMax, Murf, Hume). It is the whole workload: give it instead of monthly_pages, monthly_searches, monthly_browser_hours and monthly_emails, not with them. Only text-to-speech products can serve it; other products are out of scope.
  • needs_dedicated_ip · boolean · optionnel
    The plan must come with a dedicated sending IP address, for transactional email. A dedicated IP sold only as an add-on on top of the plan, or only on another plan, does not count.
  • needs_official_mcp · boolean · optionnel
    The provider must publish its own MCP server (hosted by it or run by the buyer), stated in its official documentation. A server only a third party published does not count.
  • ai_extraction_pages · integer · optionnel
    How many of the pages need their fields extracted by an AI model (from a prompt or a schema, or automatically) without writing selectors. Each page counts in one of the two extraction fields.
  • min_pages_per_crawl · integer · optionnel
    Pages one crawl job (from one start URL, following links) must be able to reach. A plan whose crawl has no page cap meets any number; give the size of the largest site to crawl. Different from the monthly page total.
  • min_sending_domains · integer · optionnel
    Sending domains the plan must let one account authenticate, for transactional email. Met when the stated limit is at least this many (or the pages say it is unlimited); a plan whose pages do not state a limit is unknown.
  • min_session_minutes · number · optionnel
    Minutes one browser session must be able to run on this plan, for hosted browsers. Met when the stated longest session of the plan is at least this long; a plan whose pages do not state a session limit is unknown, and a limit the page gives only for another plan is not met.
  • monthly_audio_hours · number · optionnel
    Hours of audio transcribed per month, for speech-to-text APIs (Deepgram, AssemblyAI, Speechmatics, Amazon Transcribe, Google Cloud, OpenAI). A price given per minute is multiplied by 60. It is the whole workload: give it instead of monthly_pages, monthly_searches, monthly_browser_hours, monthly_emails and monthly_characters, not with them. Only speech-to-text products can serve it; other products are out of scope.
  • monthly_pages_total · integer · optionnel
    Total pages per month when the split by type is not known yet. Costs and verdicts then cover every share of JavaScript-rendered pages from 0% to 100% of the total (no premium proxies), and the result lists monthly_pages under needs_input.
  • needs_voice_cloning · boolean · optionnel
    The plan must let the account clone a voice from its own recording, for text-to-speech. Cloning limited to eligible customers, a request form or an enterprise plan does not count for a plan that is not that plan.
  • persistent_profiles · boolean · optionnel
    The plan must keep browser profiles (login state, cookies) between sessions.
  • search_result_types · array · optionnel
    Result types the parsed search response must contain: organic, ads, local_pack, knowledge_graph, people_also_ask, related_searches, shopping, news, images, videos, ai_overview.
  • max_monthly_cost_usd · number · optionnel
    Monthly budget in USD before tax. Compared with listed prices; annual plans use their per-month equivalent.
  • needs_batch_discount · boolean · optionnel
    An official page must say a batch or asynchronous way of sending documents costs less than the list price on the plan, for document parsing. The budget and the cost are judged at the list price, not at the batch price.
  • needs_commercial_use · boolean · optionnel
    The plan must allow commercial use of the generated audio, for text-to-speech. A free plan limited to non-commercial use does not count; a plan whose pages do not say is unknown.
  • needs_no_min_billing · boolean · optionnel
    An official page must say audio is billed by the second (or an increment of seconds) with no minimum billed duration per request, for speech-to-text. A page that says nothing about rounding is unknown.
  • monthly_browser_hours · number · optionnel
    Browser time per month in hours, for hosted browsers (Browserbase, Browserless) billed by browser time. It is the whole workload: give it instead of monthly_pages, monthly_searches and browser_seconds_per_page, not with them. Only browser products can serve it; other products are out of scope.
  • max_overage_usd_per_1k · number · optionnel
    The most the plan may charge for each 1,000 emails beyond its included volume, in USD, for transactional email. A plan whose extra price is not published as a per-1,000 figure is unknown.
  • min_log_retention_days · integer · optionnel
    Days the plan must keep the logs or events of sent email where the account can read them, for transactional email. Longer retention sold as an add-on does not count.
  • monthly_document_pages · number · optionnel
    Pages of documents parsed per month, for document-parsing APIs (LlamaParse, Reducto, Unstructured, Mistral OCR, LandingAI, Amazon Textract, Google Document AI, Azure Document Intelligence, Datalab, Mathpix, Mindee, Nanonets, Upstage). A price given per credit is multiplied by the credits one page costs. It is the whole workload: give it instead of monthly_pages, monthly_searches, monthly_browser_hours, monthly_emails, monthly_characters and monthly_audio_hours, not with them. Only document-parsing products can serve it; other products are out of scope.
  • needs_table_extraction · boolean · optionnel
    The service must return the tables of a page as structure (rows and cells), not only the text, for document parsing.
  • search_returns_content · boolean · optionnel
    The search response itself must carry the text of the result pages (extracted content), not only titles, links and snippets.
  • min_requests_per_minute · number · optionnel
    Minimum request rate the plan must allow, per minute.
  • browser_seconds_per_page · number · optionnel
    Browser time per page in seconds, for headless-browser products billed by time. Every plan must then serve pages that take this long: a hosted browser does, a request-based API only when one request may run that long.
  • realtime_search_required · boolean · optionnel
    Search results must be returned in real time (not through a queue).
  • target_errors_not_billed · boolean · optionnel
    Error responses from the target site (e.g. 404) must not be billed.
  • needs_recurring_free_tier · boolean · optionnel
    The plan must be usable every month without paying (a free plan with a monthly allowance or a monthly free credit).
  • failed_requests_not_billed · boolean · optionnel
    Requests that fail on the provider side (timeouts, blocks, 5xx) must not be billed. A flat monthly price for unlimited requests meets it, since no request is billed on its own; when only success-based billing will do, use only_successful_requests_billed.
  • structured_extraction_pages · integer · optionnel
    How many of the pages must come back as JSON fields rather than HTML or text, by any method the plan offers (CSS or XPath selectors, a schema, or AI). When the fields must be found without writing selectors, use ai_extraction_pages instead; each page counts in one of the two.
  • only_successful_requests_billed · boolean · optionnel
    Billing must count successful requests only: each successful request is charged (or draws credits) and failed ones cost nothing. A flat monthly price for unlimited requests does not meet it, since the price is the same whatever succeeds.

  Exemple d'appel

  {"name": "check_conditions", "arguments": {"plans": [], "detail": "summary", "format": "full", "mcp_auth": [], "needs_zdr": true, "offer_ids": []}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}]}

• cost_scenarios≈ 19708 tokens

  Use this when: the user asks how the cost changes as the workload grows or shifts, or at what volume another plan becomes the cheapest. Do not use this when: the workload is fixed (use search). Show how API plan costs change when the monthly workload changes, for example twice the pages or a larger share of JavaScript pages. Along the path from the current to a target workload it returns the cheapest fully matching plan at both ends, each workload where the cheapest plan changes, and for tracked plans where the cost first steps up, where the plan can no longer cover the volume and where the budget is exceeded. Costs follow published pricing rules (whole packs of extra credits, per-period pack limits, per-connection browser billing, monthly minimums) with official source sentences. Example: 10,000 searches a month growing to 200,000.

  Paramètres, retour & exemple (54)

  Paramètres

  • scale · number · optionnel
    Multiply every monthly count by this factor for the end of the path, e.g. 2 for twice the current workload. Give target or scale.
  • detail · string · optionnel · summary | full
    summary returns one source sentence per fact; full returns every recorded fact and sentence.
  • target · object · optionnel
    Workload at the end of the path; fields not given keep their current value. Give target or scale.
  • mcp_auth · array · optionnel
    Ways to authenticate to the provider's hosted MCP endpoint, any one of which will do: api_key, oauth. Met when the endpoint accepts at least one of them; implies a hosted endpoint.
  • needs_zdr · boolean · optionnel
    Zero data retention must be usable on this plan. A provider that offers it only on a custom enterprise agreement does not meet it on its listed plans.
  • offer_ids · array · optionnel
    Compare only these plans and follow each of them along the path. Default: every plan that can serve the workload, following the cheapest full match at the start.
  • preferred · object · optionnel
    Nice-to-have conditions; they do not filter, they break ties.
  • needs_soc2 · boolean|string · optionnel · True | False | II
    The provider must hold a SOC 2 report: true for Type I or Type II, "II" for Type II.
  • needs_batch · boolean · optionnel
    The plan must transcribe recorded audio (a file or a stored recording), for speech-to-text. A plan priced for live audio only is not met.
  • data_regions · array · optionnel
    Regions where data must be stored or processed, e.g. ["EU"]. Different from exit-IP countries.
  • needs_korean · boolean · optionnel
    The service must voice Korean text (text-to-speech) or transcribe Korean speech (speech-to-text).
  • billing_types · array · optionnel
    Acceptable billing: monthly_subscription (A plan with a fixed fee (or minimum commitment) billed every month.) annual_subscription (A plan paid for a year at a time.) pay_as_you_go (No plan fee; a price per unit of use, billed after use or drawn from a balance topped up with any amount.) prepaid_credits (Money paid before use: a credit pack, a balance topped up first, or a plan amount charged in advance for the period.) Free plans always qualify.
  • monthly_pages · object · optionnel
    Pages per month by type: plain_html (static pages), javascript (needs rendering), premium_proxy (blocked sites that need residential/premium proxies), javascript_premium_proxy (both).
  • needs_inbound · boolean · optionnel
    The service must receive email for the buyer and hand it on (inbound parsing, routes), for transactional email.
  • product_types · array · optionnel
    Limit to these product types. Default: every type that can serve the workload.
  • search_engines · array · optionnel
    Search engines whose results the plan must return, each at the plan's search price: google, bing, yahoo, yandex, baidu, duckduckgo, naver, brave (brave is Brave's own index). monthly_searches is the total over all of them.
  • captcha_solving · boolean · optionnel
    The plan must solve CAPTCHAs for the browser sessions (the vendor does it, not the buyer own service).
  • min_concurrency · integer · optionnel
    Minimum number of concurrent requests (or browsers) the plan must allow.
  • needs_streaming · boolean · optionnel
    The service must return audio in real time as it is made (streaming or a WebSocket), for text-to-speech; for speech-to-text, it must transcribe live audio as it arrives. A speech-to-text price line is per mode, so a plan priced for recorded audio only is not met.
  • monthly_searches · integer · optionnel
    Search requests per month (one results page each).
  • needs_remote_mcp · boolean · optionnel
    The provider must host a remote MCP endpoint (a URL an MCP client connects to). An official server that can only be run on the buyer own machine does not meet it.
  • proxies_included · boolean · optionnel
    The plan must include proxy traffic of the vendor (some gigabytes in the plan fee). Proxy traffic sold only per gigabyte on top of the fee, or only on another plan, does not count.
  • search_countries · array · optionnel
    ISO 3166-1 alpha-2 codes of the countries whose search results the searches must return (the search engine's country setting, e.g. Google gl), e.g. ["GB","CA"]. Different from exit_ip_countries.
  • exit_ip_countries · array · optionnel
    ISO 3166-1 alpha-2 codes of countries the page requests must exit from, e.g. ["US","DE"]. For search results of a country use search_countries.
  • min_pages_per_doc · integer · optionnel
    Pages one document may have at the most the plan allows, for document parsing. Met when the stated limit is at least this many (or the pages say there is no page limit); a plan whose pages do not state a limit is unknown. The limit of the largest way of sending (batch or asynchronous) counts.
  • needs_diarization · boolean · optionnel
    The plan must label who spoke (speaker diarization), for speech-to-text. Where the vendor charges extra for it, the extra price is added to the monthly cost; where the pages state no price or no inclusion the plan is unknown.
  • needs_no_training · boolean · optionnel
    An official page must say the vendor does not use the buyer's text, audio or documents to train its models without the buyer opting in. A use that is on by default with an opt-out to ask for does not count.
  • session_recording · boolean · optionnel
    The plan must record browser sessions (video or replay) that the buyer can watch afterwards.
  • needs_dedicated_ip · boolean · optionnel
    The plan must come with a dedicated sending IP address, for transactional email. A dedicated IP sold only as an add-on on top of the plan, or only on another plan, does not count.
  • needs_official_mcp · boolean · optionnel
    The provider must publish its own MCP server (hosted by it or run by the buyer), stated in its official documentation. A server only a third party published does not count.
  • ai_extraction_pages · integer · optionnel
    How many of the pages need their fields extracted by an AI model (from a prompt or a schema, or automatically) without writing selectors. Each page counts in one of the two extraction fields.
  • min_pages_per_crawl · integer · optionnel
    Pages one crawl job (from one start URL, following links) must be able to reach. A plan whose crawl has no page cap meets any number; give the size of the largest site to crawl. Different from the monthly page total.
  • min_sending_domains · integer · optionnel
    Sending domains the plan must let one account authenticate, for transactional email. Met when the stated limit is at least this many (or the pages say it is unlimited); a plan whose pages do not state a limit is unknown.
  • min_session_minutes · number · optionnel
    Minutes one browser session must be able to run on this plan, for hosted browsers. Met when the stated longest session of the plan is at least this long; a plan whose pages do not state a session limit is unknown, and a limit the page gives only for another plan is not met.
  • monthly_pages_total · integer · optionnel
    Total pages per month when the split by type is not known yet. Costs and verdicts then cover every share of JavaScript-rendered pages from 0% to 100% of the total (no premium proxies), and the result lists monthly_pages under needs_input.
  • needs_voice_cloning · boolean · optionnel
    The plan must let the account clone a voice from its own recording, for text-to-speech. Cloning limited to eligible customers, a request form or an enterprise plan does not count for a plan that is not that plan.
  • persistent_profiles · boolean · optionnel
    The plan must keep browser profiles (login state, cookies) between sessions.
  • search_result_types · array · optionnel
    Result types the parsed search response must contain: organic, ads, local_pack, knowledge_graph, people_also_ask, related_searches, shopping, news, images, videos, ai_overview.
  • max_monthly_cost_usd · number · optionnel
    Monthly budget in USD before tax. Compared with listed prices; annual plans use their per-month equivalent.
  • needs_batch_discount · boolean · optionnel
    An official page must say a batch or asynchronous way of sending documents costs less than the list price on the plan, for document parsing. The budget and the cost are judged at the list price, not at the batch price.
  • needs_commercial_use · boolean · optionnel
    The plan must allow commercial use of the generated audio, for text-to-speech. A free plan limited to non-commercial use does not count; a plan whose pages do not say is unknown.
  • needs_no_min_billing · boolean · optionnel
    An official page must say audio is billed by the second (or an increment of seconds) with no minimum billed duration per request, for speech-to-text. A page that says nothing about rounding is unknown.
  • max_overage_usd_per_1k · number · optionnel
    The most the plan may charge for each 1,000 emails beyond its included volume, in USD, for transactional email. A plan whose extra price is not published as a per-1,000 figure is unknown.
  • min_log_retention_days · integer · optionnel
    Days the plan must keep the logs or events of sent email where the account can read them, for transactional email. Longer retention sold as an add-on does not count.
  • needs_table_extraction · boolean · optionnel
    The service must return the tables of a page as structure (rows and cells), not only the text, for document parsing.
  • search_returns_content · boolean · optionnel
    The search response itself must carry the text of the result pages (extracted content), not only titles, links and snippets.
  • min_requests_per_minute · number · optionnel
    Minimum request rate the plan must allow, per minute.
  • browser_seconds_per_page · number · optionnel
    Browser time per page in seconds, for headless-browser products billed by time. Every plan must then serve pages that take this long: a hosted browser does, a request-based API only when one request may run that long.
  • realtime_search_required · boolean · optionnel
    Search results must be returned in real time (not through a queue).
  • target_errors_not_billed · boolean · optionnel
    Error responses from the target site (e.g. 404) must not be billed.
  • needs_recurring_free_tier · boolean · optionnel
    The plan must be usable every month without paying (a free plan with a monthly allowance or a monthly free credit).
  • failed_requests_not_billed · boolean · optionnel
    Requests that fail on the provider side (timeouts, blocks, 5xx) must not be billed. A flat monthly price for unlimited requests meets it, since no request is billed on its own; when only success-based billing will do, use only_successful_requests_billed.
  • structured_extraction_pages · integer · optionnel
    How many of the pages must come back as JSON fields rather than HTML or text, by any method the plan offers (CSS or XPath selectors, a schema, or AI). When the fields must be found without writing selectors, use ai_extraction_pages instead; each page counts in one of the two.
  • only_successful_requests_billed · boolean · optionnel
    Billing must count successful requests only: each successful request is charged (or draws credits) and failed ones cost nothing. A flat monthly price for unlimited requests does not meet it, since the price is the same whatever succeeds.

  Exemple d'appel

  {"name": "cost_scenarios", "arguments": {"scale": 1, "detail": "summary", "target": {}, "mcp_auth": [], "needs_zdr": true, "offer_ids": []}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"path": {}, "as_of": "texte", "notes": [], "plans": [], "at_end": [], "counts": {}, "request": {}, "tracked": [], "at_start": [], "compared": "all_plans", "switches": [], "link_notes": {}}}

• get_details≈ 2646 tokens

  Use this when: you need every recorded condition, pricing rule and source sentence of one plan or product found by search or named by the user. Do not use this when: you want to compare plans against a workload (use search or check_conditions). Get every recorded condition for one API plan or product: value, which plans it applies to, exceptions, official source sentence, check date and purchase link. A product too long for one response comes with one source sentence per fact. Example: every recorded condition of 1 plan found by search.

  Paramètres, retour & exemple (2)

  Paramètres

  • offer_id · string · optionnel
    A plan id, e.g. "firecrawl-standard".
  • product_id · string · optionnel
    A product id, e.g. "firecrawl-api".

  Exemple d'appel

  {"name": "get_details", "arguments": {"offer_id": "texte", "product_id": "texte"}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"as_of": "texte", "offers": [], "product": {}, "provider": {}, "link_notes": {}, "evidence_detail": "summary"}}

• relax_conditions≈ 11194 tokens

  Use this when: search finds no plan or too few plans and the user wants to know which condition to loosen and by how much. Do not use this when: plans already match (use search or check_conditions). When no API plan, or too few API plans, meet every required condition: for each condition the user can change, returns the smallest change that lets at least one plan meet everything (for example the monthly budget needed, the concurrency available, or one exit-IP country to drop) and the plans it admits with their monthly cost. Also lists plans that fail only because a condition is not stated on the official pages, with the pages checked and the check date, so the user can confirm with the provider. Example: no plan meets 5,000 hours of audio a month under $100.

  Paramètres, retour & exemple (53)

  Paramètres

  • limit · integer · optionnel
    Maximum plans listed for each change and under confirm_with_provider (default 3).
  • detail · string · optionnel · summary | full
    summary returns one source sentence per fact; full returns every recorded fact and sentence.
  • mcp_auth · array · optionnel
    Ways to authenticate to the provider's hosted MCP endpoint, any one of which will do: api_key, oauth. Met when the endpoint accepts at least one of them; implies a hosted endpoint.
  • needs_zdr · boolean · optionnel
    Zero data retention must be usable on this plan. A provider that offers it only on a custom enterprise agreement does not meet it on its listed plans.
  • preferred · object · optionnel
    Nice-to-have conditions; they do not filter, they break ties.
  • needs_soc2 · boolean|string · optionnel · True | False | II
    The provider must hold a SOC 2 report: true for Type I or Type II, "II" for Type II.
  • negotiable · array · optionnel
    Conditions the user said can change: monthly_volume, concurrency, rate_limit, monthly_budget, billing_type, free_allowance, failed_request_billing, success_billing, target_error_billing, geo_targeting, data_region, serp_response_mode, serp_localization, serp_result_types, serp_engines, crawl_page_limit, mcp_official, mcp_remote, mcp_auth, zdr, soc2, search_content, min_session_minutes, proxies_included, captcha_solving, session_recording, persistent_profiles, dedicated_ip, sending_domains, log_retention_days, inbound_parse, overage_per_1k, commercial_use, voice_cloning, streaming, languages_ko, data_not_used_for_training, batch, diarization, no_min_billing, batch_discount, max_pages_per_doc, table_extraction. Default: every required condition, each marked ask_user because the user has not said it can change.
  • needs_batch · boolean · optionnel
    The plan must transcribe recorded audio (a file or a stored recording), for speech-to-text. A plan priced for live audio only is not met.
  • data_regions · array · optionnel
    Regions where data must be stored or processed, e.g. ["EU"]. Different from exit-IP countries.
  • needs_korean · boolean · optionnel
    The service must voice Korean text (text-to-speech) or transcribe Korean speech (speech-to-text).
  • billing_types · array · optionnel
    Acceptable billing: monthly_subscription (A plan with a fixed fee (or minimum commitment) billed every month.) annual_subscription (A plan paid for a year at a time.) pay_as_you_go (No plan fee; a price per unit of use, billed after use or drawn from a balance topped up with any amount.) prepaid_credits (Money paid before use: a credit pack, a balance topped up first, or a plan amount charged in advance for the period.) Free plans always qualify.
  • monthly_pages · object · optionnel
    Pages per month by type: plain_html (static pages), javascript (needs rendering), premium_proxy (blocked sites that need residential/premium proxies), javascript_premium_proxy (both).
  • needs_inbound · boolean · optionnel
    The service must receive email for the buyer and hand it on (inbound parsing, routes), for transactional email.
  • product_types · array · optionnel
    Limit to these product types. Default: every type that can serve the workload.
  • search_engines · array · optionnel
    Search engines whose results the plan must return, each at the plan's search price: google, bing, yahoo, yandex, baidu, duckduckgo, naver, brave (brave is Brave's own index). monthly_searches is the total over all of them.
  • captcha_solving · boolean · optionnel
    The plan must solve CAPTCHAs for the browser sessions (the vendor does it, not the buyer own service).
  • min_concurrency · integer · optionnel
    Minimum number of concurrent requests (or browsers) the plan must allow.
  • needs_streaming · boolean · optionnel
    The service must return audio in real time as it is made (streaming or a WebSocket), for text-to-speech; for speech-to-text, it must transcribe live audio as it arrives. A speech-to-text price line is per mode, so a plan priced for recorded audio only is not met.
  • monthly_searches · integer · optionnel
    Search requests per month (one results page each).
  • needs_remote_mcp · boolean · optionnel
    The provider must host a remote MCP endpoint (a URL an MCP client connects to). An official server that can only be run on the buyer own machine does not meet it.
  • proxies_included · boolean · optionnel
    The plan must include proxy traffic of the vendor (some gigabytes in the plan fee). Proxy traffic sold only per gigabyte on top of the fee, or only on another plan, does not count.
  • search_countries · array · optionnel
    ISO 3166-1 alpha-2 codes of the countries whose search results the searches must return (the search engine's country setting, e.g. Google gl), e.g. ["GB","CA"]. Different from exit_ip_countries.
  • exit_ip_countries · array · optionnel
    ISO 3166-1 alpha-2 codes of countries the page requests must exit from, e.g. ["US","DE"]. For search results of a country use search_countries.
  • min_pages_per_doc · integer · optionnel
    Pages one document may have at the most the plan allows, for document parsing. Met when the stated limit is at least this many (or the pages say there is no page limit); a plan whose pages do not state a limit is unknown. The limit of the largest way of sending (batch or asynchronous) counts.
  • needs_diarization · boolean · optionnel
    The plan must label who spoke (speaker diarization), for speech-to-text. Where the vendor charges extra for it, the extra price is added to the monthly cost; where the pages state no price or no inclusion the plan is unknown.
  • needs_no_training · boolean · optionnel
    An official page must say the vendor does not use the buyer's text, audio or documents to train its models without the buyer opting in. A use that is on by default with an opt-out to ask for does not count.
  • session_recording · boolean · optionnel
    The plan must record browser sessions (video or replay) that the buyer can watch afterwards.
  • needs_dedicated_ip · boolean · optionnel
    The plan must come with a dedicated sending IP address, for transactional email. A dedicated IP sold only as an add-on on top of the plan, or only on another plan, does not count.
  • needs_official_mcp · boolean · optionnel
    The provider must publish its own MCP server (hosted by it or run by the buyer), stated in its official documentation. A server only a third party published does not count.
  • ai_extraction_pages · integer · optionnel
    How many of the pages need their fields extracted by an AI model (from a prompt or a schema, or automatically) without writing selectors. Each page counts in one of the two extraction fields.
  • min_pages_per_crawl · integer · optionnel
    Pages one crawl job (from one start URL, following links) must be able to reach. A plan whose crawl has no page cap meets any number; give the size of the largest site to crawl. Different from the monthly page total.
  • min_sending_domains · integer · optionnel
    Sending domains the plan must let one account authenticate, for transactional email. Met when the stated limit is at least this many (or the pages say it is unlimited); a plan whose pages do not state a limit is unknown.
  • min_session_minutes · number · optionnel
    Minutes one browser session must be able to run on this plan, for hosted browsers. Met when the stated longest session of the plan is at least this long; a plan whose pages do not state a session limit is unknown, and a limit the page gives only for another plan is not met.
  • monthly_pages_total · integer · optionnel
    Total pages per month when the split by type is not known yet. Costs and verdicts then cover every share of JavaScript-rendered pages from 0% to 100% of the total (no premium proxies), and the result lists monthly_pages under needs_input.
  • needs_voice_cloning · boolean · optionnel
    The plan must let the account clone a voice from its own recording, for text-to-speech. Cloning limited to eligible customers, a request form or an enterprise plan does not count for a plan that is not that plan.
  • persistent_profiles · boolean · optionnel
    The plan must keep browser profiles (login state, cookies) between sessions.
  • search_result_types · array · optionnel
    Result types the parsed search response must contain: organic, ads, local_pack, knowledge_graph, people_also_ask, related_searches, shopping, news, images, videos, ai_overview.
  • max_monthly_cost_usd · number · optionnel
    Monthly budget in USD before tax. Compared with listed prices; annual plans use their per-month equivalent.
  • needs_batch_discount · boolean · optionnel
    An official page must say a batch or asynchronous way of sending documents costs less than the list price on the plan, for document parsing. The budget and the cost are judged at the list price, not at the batch price.
  • needs_commercial_use · boolean · optionnel
    The plan must allow commercial use of the generated audio, for text-to-speech. A free plan limited to non-commercial use does not count; a plan whose pages do not say is unknown.
  • needs_no_min_billing · boolean · optionnel
    An official page must say audio is billed by the second (or an increment of seconds) with no minimum billed duration per request, for speech-to-text. A page that says nothing about rounding is unknown.
  • max_overage_usd_per_1k · number · optionnel
    The most the plan may charge for each 1,000 emails beyond its included volume, in USD, for transactional email. A plan whose extra price is not published as a per-1,000 figure is unknown.
  • min_log_retention_days · integer · optionnel
    Days the plan must keep the logs or events of sent email where the account can read them, for transactional email. Longer retention sold as an add-on does not count.
  • needs_table_extraction · boolean · optionnel
    The service must return the tables of a page as structure (rows and cells), not only the text, for document parsing.
  • search_returns_content · boolean · optionnel
    The search response itself must carry the text of the result pages (extracted content), not only titles, links and snippets.
  • min_requests_per_minute · number · optionnel
    Minimum request rate the plan must allow, per minute.
  • browser_seconds_per_page · number · optionnel
    Browser time per page in seconds, for headless-browser products billed by time. Every plan must then serve pages that take this long: a hosted browser does, a request-based API only when one request may run that long.
  • realtime_search_required · boolean · optionnel
    Search results must be returned in real time (not through a queue).
  • target_errors_not_billed · boolean · optionnel
    Error responses from the target site (e.g. 404) must not be billed.
  • needs_recurring_free_tier · boolean · optionnel
    The plan must be usable every month without paying (a free plan with a monthly allowance or a monthly free credit).
  • failed_requests_not_billed · boolean · optionnel
    Requests that fail on the provider side (timeouts, blocks, 5xx) must not be billed. A flat monthly price for unlimited requests meets it, since no request is billed on its own; when only success-based billing will do, use only_successful_requests_billed.
  • structured_extraction_pages · integer · optionnel
    How many of the pages must come back as JSON fields rather than HTML or text, by any method the plan offers (CSS or XPath selectors, a schema, or AI). When the fields must be found without writing selectors, use ai_extraction_pages instead; each page counts in one of the two.
  • only_successful_requests_billed · boolean · optionnel
    Billing must count successful requests only: each successful request is charged (or draws credits) and failed ones cost nothing. A flat monthly price for unlimited requests does not meet it, since the price is the same whatever succeeds.

  Exemple d'appel

  {"name": "relax_conditions", "arguments": {"limit": 1, "detail": "summary", "mcp_auth": [], "needs_zdr": true, "preferred": {}, "needs_soc2": true}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"as_of": "texte", "notes": [], "counts": {}, "changes": [], "request": {}, "link_notes": {}, "assumptions": [], "needs_input": [], "confirm_with_provider": {}}}

• search≈ 15544 tokens

  Call this right before choosing or buying an API plan: it returns, for each plan, whether it meets every condition, the monthly cost, and the official sentence behind each verdict. Use this when: the user wants to know which API plans (web scraping, SERP, web search, browser infrastructure, transactional email, text-to-speech, speech-to-text, document parsing) fit a monthly workload, budget or required conditions, or what one would cost per month at that workload. Do not use this when: the user wants to scrape, search the web, send email, generate speech, transcribe audio or parse a document itself (this tool reads no pages and calls no provider), needs proxies only, or asks about an API outside these fields. Call it right before you choose or buy a tool or plan; format "brief" returns a short answer. Find API plans (web scraping, SERP, web search, browser infrastructure, transactional email, text-to-speech, speech-to-text, document parsing) that meet required conditions: monthly pages by type (plain HTML, JavaScript rendering, blocked sites that need premium proxies), monthly searches, concurrency, requests per minute, monthly budget before tax, billing type, exit-IP countries, search engines, search-result countries and parsed result types, pages per crawl, data storage region and structured (JSON) or AI extraction. Browser infrastructure is asked in monthly browser hours, transactional email in monthly emails, text-to-speech in monthly characters, speech-to-text in monthly audio hours and document parsing in monthly document pages, each with its own conditions (for example session length, dedicated IP, voice cloning, Korean, live or recorded audio, speaker labels, pages per document, table extraction, batch discount). For search APIs it also checks the agent connection: an official or remote (hosted) MCP server and how it signs in (API key or OAuth), zero data retention on the plan, a SOC 2 report, and whether results carry the page text. Returns each plan's monthly cost, whether it meets every required condition, and for every condition met / not met / unknown with the official source sentence and check date, plus a purchase link. Plans with an unknown required condition are listed separately as partial matches. If nothing matches, returns how many plans each condition excluded and the cheapest plan that meets all the other conditions. If you already have candidate plans or providers (for example from a web search), pass them as providers (search) or offer_ids (check_conditions) so that only those are judged. Hosts that support MCP Apps show the result as a comparison card. Example: 50,000 SERP searches a month in GB under $100; 300 hours of Korean audio a month with speaker labels.

  Paramètres, retour & exemple (59)

  Paramètres

  • limit · integer · optionnel
    Maximum plans returned in each list (default and maximum 5).
  • detail · string · optionnel · summary | full
    summary returns one source sentence per fact; full returns every recorded fact and sentence.
  • format · string · optionnel · full | brief
    brief returns, per plan, the plan, provider, whether it meets all, the monthly cost and the requested conditions only, each with its verdict, a one-line reason, one source sentence and its source (about a quarter of the full size); the verdicts and costs are those of full. Call it right before choosing or buying a plan.
  • mcp_auth · array · optionnel
    Ways to authenticate to the provider's hosted MCP endpoint, any one of which will do: api_key, oauth. Met when the endpoint accepts at least one of them; implies a hosted endpoint.
  • needs_zdr · boolean · optionnel
    Zero data retention must be usable on this plan. A provider that offers it only on a custom enterprise agreement does not meet it on its listed plans.
  • preferred · object · optionnel
    Nice-to-have conditions; they do not filter, they break ties.
  • providers · array · optionnel
    Judge only the plans of these providers (candidates you already have, for example from a web search). Give provider ids or names as listed, such as "tavily", "Brave Search API" or "You.com"; case does not matter. Names that are not providers here are listed under unknown_providers in the result.
  • needs_soc2 · boolean|string · optionnel · True | False | II
    The provider must hold a SOC 2 report: true for Type I or Type II, "II" for Type II.
  • needs_batch · boolean · optionnel
    The plan must transcribe recorded audio (a file or a stored recording), for speech-to-text. A plan priced for live audio only is not met.
  • data_regions · array · optionnel
    Regions where data must be stored or processed, e.g. ["EU"]. Different from exit-IP countries.
  • needs_korean · boolean · optionnel
    The service must voice Korean text (text-to-speech) or transcribe Korean speech (speech-to-text).
  • billing_types · array · optionnel
    Acceptable billing: monthly_subscription (A plan with a fixed fee (or minimum commitment) billed every month.) annual_subscription (A plan paid for a year at a time.) pay_as_you_go (No plan fee; a price per unit of use, billed after use or drawn from a balance topped up with any amount.) prepaid_credits (Money paid before use: a credit pack, a balance topped up first, or a plan amount charged in advance for the period.) Free plans always qualify.
  • monthly_pages · object · optionnel
    Pages per month by type: plain_html (static pages), javascript (needs rendering), premium_proxy (blocked sites that need residential/premium proxies), javascript_premium_proxy (both).
  • needs_inbound · boolean · optionnel
    The service must receive email for the buyer and hand it on (inbound parsing, routes), for transactional email.
  • product_types · array · optionnel
    Limit to these product types. Default: every type that can serve the workload.
  • monthly_emails · number · optionnel
    Emails sent per month, for transactional email APIs (SendGrid, Mailgun, Postmark, Resend, Amazon SES, Brevo, Mailtrap, Loops). It is the whole workload: give it instead of monthly_pages, monthly_searches and monthly_browser_hours, not with them. Only email products can serve it; other products are out of scope.
  • search_engines · array · optionnel
    Search engines whose results the plan must return, each at the plan's search price: google, bing, yahoo, yandex, baidu, duckduckgo, naver, brave (brave is Brave's own index). monthly_searches is the total over all of them.
  • captcha_solving · boolean · optionnel
    The plan must solve CAPTCHAs for the browser sessions (the vendor does it, not the buyer own service).
  • min_concurrency · integer · optionnel
    Minimum number of concurrent requests (or browsers) the plan must allow.
  • needs_streaming · boolean · optionnel
    The service must return audio in real time as it is made (streaming or a WebSocket), for text-to-speech; for speech-to-text, it must transcribe live audio as it arrives. A speech-to-text price line is per mode, so a plan priced for recorded audio only is not met.
  • monthly_searches · integer · optionnel
    Search requests per month (one results page each).
  • needs_remote_mcp · boolean · optionnel
    The provider must host a remote MCP endpoint (a URL an MCP client connects to). An official server that can only be run on the buyer own machine does not meet it.
  • proxies_included · boolean · optionnel
    The plan must include proxy traffic of the vendor (some gigabytes in the plan fee). Proxy traffic sold only per gigabyte on top of the fee, or only on another plan, does not count.
  • search_countries · array · optionnel
    ISO 3166-1 alpha-2 codes of the countries whose search results the searches must return (the search engine's country setting, e.g. Google gl), e.g. ["GB","CA"]. Different from exit_ip_countries.
  • exit_ip_countries · array · optionnel
    ISO 3166-1 alpha-2 codes of countries the page requests must exit from, e.g. ["US","DE"]. For search results of a country use search_countries.
  • min_pages_per_doc · integer · optionnel
    Pages one document may have at the most the plan allows, for document parsing. Met when the stated limit is at least this many (or the pages say there is no page limit); a plan whose pages do not state a limit is unknown. The limit of the largest way of sending (batch or asynchronous) counts.
  • needs_diarization · boolean · optionnel
    The plan must label who spoke (speaker diarization), for speech-to-text. Where the vendor charges extra for it, the extra price is added to the monthly cost; where the pages state no price or no inclusion the plan is unknown.
  • needs_no_training · boolean · optionnel
    An official page must say the vendor does not use the buyer's text, audio or documents to train its models without the buyer opting in. A use that is on by default with an opt-out to ask for does not count.
  • session_recording · boolean · optionnel
    The plan must record browser sessions (video or replay) that the buyer can watch afterwards.
  • monthly_characters · number · optionnel
    Characters of text voiced per month, for text-to-speech APIs (ElevenLabs, Cartesia, Deepgram Aura, OpenAI, Google Cloud, Amazon Polly, Azure, Gradium, Rime, Inworld, MiniMax, Murf, Hume). It is the whole workload: give it instead of monthly_pages, monthly_searches, monthly_browser_hours and monthly_emails, not with them. Only text-to-speech products can serve it; other products are out of scope.
  • needs_dedicated_ip · boolean · optionnel
    The plan must come with a dedicated sending IP address, for transactional email. A dedicated IP sold only as an add-on on top of the plan, or only on another plan, does not count.
  • needs_official_mcp · boolean · optionnel
    The provider must publish its own MCP server (hosted by it or run by the buyer), stated in its official documentation. A server only a third party published does not count.
  • ai_extraction_pages · integer · optionnel
    How many of the pages need their fields extracted by an AI model (from a prompt or a schema, or automatically) without writing selectors. Each page counts in one of the two extraction fields.
  • min_pages_per_crawl · integer · optionnel
    Pages one crawl job (from one start URL, following links) must be able to reach. A plan whose crawl has no page cap meets any number; give the size of the largest site to crawl. Different from the monthly page total.
  • min_sending_domains · integer · optionnel
    Sending domains the plan must let one account authenticate, for transactional email. Met when the stated limit is at least this many (or the pages say it is unlimited); a plan whose pages do not state a limit is unknown.
  • min_session_minutes · number · optionnel
    Minutes one browser session must be able to run on this plan, for hosted browsers. Met when the stated longest session of the plan is at least this long; a plan whose pages do not state a session limit is unknown, and a limit the page gives only for another plan is not met.
  • monthly_audio_hours · number · optionnel
    Hours of audio transcribed per month, for speech-to-text APIs (Deepgram, AssemblyAI, Speechmatics, Amazon Transcribe, Google Cloud, OpenAI). A price given per minute is multiplied by 60. It is the whole workload: give it instead of monthly_pages, monthly_searches, monthly_browser_hours, monthly_emails and monthly_characters, not with them. Only speech-to-text products can serve it; other products are out of scope.
  • monthly_pages_total · integer · optionnel
    Total pages per month when the split by type is not known yet. Costs and verdicts then cover every share of JavaScript-rendered pages from 0% to 100% of the total (no premium proxies), and the result lists monthly_pages under needs_input.
  • needs_voice_cloning · boolean · optionnel
    The plan must let the account clone a voice from its own recording, for text-to-speech. Cloning limited to eligible customers, a request form or an enterprise plan does not count for a plan that is not that plan.
  • persistent_profiles · boolean · optionnel
    The plan must keep browser profiles (login state, cookies) between sessions.
  • search_result_types · array · optionnel
    Result types the parsed search response must contain: organic, ads, local_pack, knowledge_graph, people_also_ask, related_searches, shopping, news, images, videos, ai_overview.
  • max_monthly_cost_usd · number · optionnel
    Monthly budget in USD before tax. Compared with listed prices; annual plans use their per-month equivalent.
  • needs_batch_discount · boolean · optionnel
    An official page must say a batch or asynchronous way of sending documents costs less than the list price on the plan, for document parsing. The budget and the cost are judged at the list price, not at the batch price.
  • needs_commercial_use · boolean · optionnel
    The plan must allow commercial use of the generated audio, for text-to-speech. A free plan limited to non-commercial use does not count; a plan whose pages do not say is unknown.
  • needs_no_min_billing · boolean · optionnel
    An official page must say audio is billed by the second (or an increment of seconds) with no minimum billed duration per request, for speech-to-text. A page that says nothing about rounding is unknown.
  • monthly_browser_hours · number · optionnel
    Browser time per month in hours, for hosted browsers (Browserbase, Browserless) billed by browser time. It is the whole workload: give it instead of monthly_pages, monthly_searches and browser_seconds_per_page, not with them. Only browser products can serve it; other products are out of scope.
  • max_overage_usd_per_1k · number · optionnel
    The most the plan may charge for each 1,000 emails beyond its included volume, in USD, for transactional email. A plan whose extra price is not published as a per-1,000 figure is unknown.
  • min_log_retention_days · integer · optionnel
    Days the plan must keep the logs or events of sent email where the account can read them, for transactional email. Longer retention sold as an add-on does not count.
  • monthly_document_pages · number · optionnel
    Pages of documents parsed per month, for document-parsing APIs (LlamaParse, Reducto, Unstructured, Mistral OCR, LandingAI, Amazon Textract, Google Document AI, Azure Document Intelligence, Datalab, Mathpix, Mindee, Nanonets, Upstage). A price given per credit is multiplied by the credits one page costs. It is the whole workload: give it instead of monthly_pages, monthly_searches, monthly_browser_hours, monthly_emails, monthly_characters and monthly_audio_hours, not with them. Only document-parsing products can serve it; other products are out of scope.
  • needs_table_extraction · boolean · optionnel
    The service must return the tables of a page as structure (rows and cells), not only the text, for document parsing.
  • search_returns_content · boolean · optionnel
    The search response itself must carry the text of the result pages (extracted content), not only titles, links and snippets.
  • min_requests_per_minute · number · optionnel
    Minimum request rate the plan must allow, per minute.
  • browser_seconds_per_page · number · optionnel
    Browser time per page in seconds, for headless-browser products billed by time. Every plan must then serve pages that take this long: a hosted browser does, a request-based API only when one request may run that long.
  • realtime_search_required · boolean · optionnel
    Search results must be returned in real time (not through a queue).
  • target_errors_not_billed · boolean · optionnel
    Error responses from the target site (e.g. 404) must not be billed.
  • needs_recurring_free_tier · boolean · optionnel
    The plan must be usable every month without paying (a free plan with a monthly allowance or a monthly free credit).
  • failed_requests_not_billed · boolean · optionnel
    Requests that fail on the provider side (timeouts, blocks, 5xx) must not be billed. A flat monthly price for unlimited requests meets it, since no request is billed on its own; when only success-based billing will do, use only_successful_requests_billed.
  • structured_extraction_pages · integer · optionnel
    How many of the pages must come back as JSON fields rather than HTML or text, by any method the plan offers (CSS or XPath selectors, a schema, or AI). When the fields must be found without writing selectors, use ai_extraction_pages instead; each page counts in one of the two.
  • only_successful_requests_billed · boolean · optionnel
    Billing must count successful requests only: each successful request is charged (or draws credits) and failed ones cost nothing. A flat monthly price for unlimited requests does not meet it, since the price is the same whatever succeeds.

  Exemple d'appel

  {"name": "search", "arguments": {"limit": 1, "detail": "summary", "format": "full", "mcp_auth": [], "needs_zdr": true, "preferred": {}}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}]}

• verify_answer≈ 11166 tokens

  Use this when: you have drafted an answer that names API plans, costs or whether they meet the user's conditions, before you give it. Do not use this when: you have not looked up any plan yet (use search). Check what a drafted answer says about API plans against the recorded official facts, using the user's conditions as given in request. For each plan claim (plan id, name or pricing-page URL, with the stated billing, monthly cost, whether it meets every condition, single condition verdicts and recorded values) it returns conflicts with the official source sentences, the check date and a one-line fix: a plan not in the catalog, unmet or unknown conditions stated as met, met stated as unmet, a condition verdict, a cost outside the computed cost or range, or a value that differs. It also lists verdicts that rest on a derived value, official pages that disagree or a two-value price range (state the condition), plans it could not check (an ambiguous name returns the candidates), the cheapest plan that meets every condition when the answer does not name it, and words of the question that suggest a condition the request does not carry. Example: a drafted answer says 1 plan costs $49 a month for 100,000 emails.

  Paramètres, retour & exemple (3)

  Paramètres

  • claims · array · requis
    One entry per plan the answer names, with what the answer says about it.
  • request · object · requis
    The user's workload and required conditions as you understood them, in the search input format. Each claim is judged against these, not against earlier calls.
  • question · string · optionnel
    The user's words, if available. Read only for words that suggest a condition the request does not carry.

  Exemple d'appel

  {"name": "verify_answer", "arguments": {"claims": [], "request": {}}}

  Exemple de retour

  {"content": [{"type": "text", "text": "…"}], "structuredContent": {"limits": [], "omitted": [], "evidence": {}, "conflicts": [], "unchecked": [], "consistent": true, "conditional": [], "request_used": {}, "checked_scope": "texte", "request_may_omit": []}}

C'est votre serveur ?

Revendiquez-le sur l'espace créateur (preuve GitHub, domaine ou npm) pour afficher le badge « propriétaire vérifié » et lancer un pentest comportemental à la demande afin d'obtenir le badge « Vérifié ».

Créer un compte / revendiquer →

Source

https://plans.intoperson.com/mcp