MCP-сервер Foglift

• Адресhttps://foglift.io/api/mcpнет в официальном реестре: ответил на запрос

Ни одна запись официального реестра MCP не называет этот сервер. Он ответил, когда Атлас запросил foglift.io по одному из обычных для MCP-серверов адресов, значит, сервер принадлежит самому вендору. Клиенты и каталоги, которые берут серверы из реестра, его не показывают.

• Состояниеотвечаетпроверено 2026-10-27
• Инструменты30
• Сервер называет себяfoglift
• Версия сервера1.1.0
• Версия протокола2025-03-26

2026-10-27 сервер ответил на initialize и tools/list без входа. Первый ответ за 118 мс.

• Делают ли инструменты работу продукта, ещё не проверено.
• Проверок пока 1 из 7, нужных, чтобы судить о надёжности.

Вывод сделан правилами по замерам выше. Вероятный первый вызов выбрала модель по списку инструментов: прочитайте описание, прежде чем на него полагаться.

• Оценка INITE MCPчерез неделюдля оценки нужно 7 ежедневных проверок, пока 1
• Отвечалидёт замерпроверок пока 1, долю покажем после 7
• Медиана первого ответа146 мс
• Место в контексте агента≈ 5 602 токеновописания всех инструментов (30), они занимают контекст агента на каждом ходу
• Описания инструментовA · 100 из 100
• Типизированные ответы0 из 30 инструментовобъявлен outputSchema: ответ приходит с типами и читается сразу

• Все инструменты проходят все шесть проверок.

Программа проверяет сервер раз в день: initialize и tools/list, без модели. Токены - это длина описаний из расчёта примерно четыре символа на токен. Оценка - среднее шести проверок имени, описания, схемы и аннотаций каждого инструмента. Она говорит, насколько понятно описаны инструменты, а не о том, работают ли они. Оценка складывается только из замеренного: отвечает ли сервер (25%), делает ли работу продукта (25%), описания (20%), контекст (10%), использование (10%) и насколько просто подключиться (10%). Незамеренная часть не учитывается, а скрытая инструкция ограничивает оценку 40 баллами. Состояние MCP-серверов вендоров: методика и цифры →

• scan_website

  Scan a website URL for SEO, GEO, and AI search readiness. Returns overall score, category scores (technical SEO, content quality, AI optimization, schema markup), and a list of specific issues with fix suggestions.

• batch_scan

  Scan multiple URLs in one request (max 10). Returns scores and issues for each URL. Requires API key.

• run_ai_visibility

  Run an AI visibility check — query AI models (ChatGPT, Claude, Perplexity, Gemini) with prompts related to your domain and see if they mention or recommend your brand. Returns per-model results with citation status and sentiment.

• get_ai_results

  Get historical AI visibility results for your domain. Shows how AI models have mentioned your brand over time, with sentiment and citation data.

• get_search_trace

  Get the ordered ChatGPT search trace for one AI visibility result, including search steps, queries, pages read, citation flags, run rollups, and the trailing 30-day prompt rollup when available.

• get_prompts

  Get your saved AI visibility monitoring prompts. These are the prompts used to check if AI models mention your brand.

• add_prompt

  Add a new prompt to monitor for AI visibility. The prompt will be periodically tested against AI models to check if they mention your brand.

• delete_prompt

  Delete a saved AI visibility monitoring prompt by its ID.

• get_models

  Get your enabled AI models and monitoring frequency settings. Account-level setting — applies to every brand on your account.

• set_models

  Update which AI models are enabled for monitoring and the monitoring frequency. Account-level setting — applies to every brand on your account.

• get_sentiment

  Get AI sentiment analysis results — shows how AI models talk about your brand (positive, neutral, negative) with trends over time.

• get_usage

  Get API usage statistics for your Foglift account — scan count, AI visibility checks, and quota remaining.

• get_scan_history

  Get historical scan results for a specific URL — shows how scores have changed over time.

• get_geo_monitor

  Get GEO monitoring data — track how your site's AI search optimization scores change over time.

• get_referrer_analytics

  Get AI-engine referrer analytics — total visits, per-engine breakdown, top landing paths, and daily trend for visitors arriving from ChatGPT, Claude, Perplexity, Gemini, Copilot, NotebookLM, and other AI engines.

• get_crawler_analytics

  Get observed AI crawler activity for a brand, including total visits, engine and crawler-category breakdowns, distinct pages reached, top paths, and the time trend. Use this for server-observed bot access; use get_referrer_analytics for…

• get_competitors

  Compare this brand with its named competitors using current and previous monitoring windows. Returns result mention rates and percentage-point changes, with incomplete rates omitted when an evidence window reaches its cap. Use this for…

• get_source_gap

  Find domains AI answers cite when this brand is absent, including which tracked competitors appear beside each source and an authority-based attainability signal. When authority is queued for refresh, authority and attainability remain…

• get_aeo_score

  Read the latest stored AI Readiness page score for a brand plus its score history and direction of change. This uses existing deterministic score evidence and does not run a new analysis or spend AI tokens.

• get_search_console

  Fetch Google Search Console or Bing Webmaster Tools data for the connected domain — top queries, top pages, and search demand with low AI visibility. Defaults to Google for backward compatibility.

• estimate_cost

  Estimate token cost before running run_ai_visibility. Accepts the same prompt/model shape as run_ai_visibility plus brand_id. Returns estimated_tokens, current_balance, and sufficient so agents can avoid overdrawing the account before a…

• get_recommendations

  Get AI visibility recommendation diagnostics for a Foglift brand without spending Foglift server-side recommendation tokens. Returns weak engines, weak prompts, citation/competitor gaps, already stored Foglift or customer-agent…

• submit_recommendation

  Submit a recommendation generated by your agent back to Foglift. Foglift stores it as source='customer_agent' for the brand and surfaces it in the dashboard and recommendation digest as generated by your agent. Include the fix, target…

• apply_recommendation

  Prepare the next action for a stored Foglift or customer-agent recommendation. For customer-site edits, returns a precise change-set instead of modifying the site. After applying, re-run run_ai_visibility and get_ai_results to verify…

• generate_content_brief

  Generate a content brief for a page in the target Foglift brand, using the brand context when brand_id is provided. Use this for recommendation follow-through on content_gap or visibility_boost fixes. Legacy workspace_id is accepted as an…

• report_issue

  Report a bug, unexpected behavior, question, or feature request to the Foglift team. You can describe this far more precisely than a human user - please do. Include exactly what you were trying to accomplish, the precise Foglift…

• whoami

  Return the authenticated Foglift account identity: user_id, plan, default workspace (name + domain), current token balance, and monthly allocation. Use this to confirm which account an API key is tied to.

• recommend_setup

  Calculate the sustainable monitoring prompt count from the authenticated Foglift account's persisted plan, active brands, scan frequency, and enabled engines. Returns the trusted prompt budget, category distribution, examples, and…

• apply_setup_recommendation

  Create the categorized monitoring prompts from a recommend_setup result for a specific Foglift brand. Requires brand_id, brand name, competitors, industry, and the recommendation output. Idempotent on retry. Legacy workspace_id is…

• recipe_run

  Run a named Foglift recipe against a specific brand. First available recipe: weekly-brand-health-report, which returns a weekly AI Readiness / AI Visibility report from existing scan and monitoring data.