# GEOly Docs ## Documentation ### Get started - [Welcome to GEOly](https://docs.geoly.ai/index.md): The GEO data platform that shows how AI engines mention, rank, cite and recommend your brand. - [Create your GEOly account and start monitoring](https://docs.geoly.ai/get-started/register-account.md): Register with Google or email, confirm your organization, and start your first GEOly monitoring workflow. - [What‘s the GEOly?](https://docs.geoly.ai/get-started/geoly-platform.md) ### Monitoring - [Prompts](https://docs.geoly.ai/analytics/geoly-prompts.md) - [Topic and Platform Analysis](https://docs.geoly.ai/analytics/ai-engine-topic.md) ### Insights - [Public Explore and Research Views](https://docs.geoly.ai/analytics/ai-engine-insight.md) - [Explore: AI Market & Brand Research](https://docs.geoly.ai/analytics/ai-engine-explore.md): Use GEOly Explore to discover public AI industry signals across categories, Topics, brands, citations, and opportunities, then turn them into research hypotheses and validate them with first-party monitoring data. - [Shopping](https://docs.geoly.ai/analytics/ai-engine-shopping.md) ### Shopify app - [How to create an llms.txt file for my website](https://docs.geoly.ai/shopify/ai-llms-txt.md) - [What is Catalog Optimization](https://docs.geoly.ai/shopify/catalog-optimization.md): Unlike traditional SEO, which focuses on keyword rankings, GEO optimizes for semantic intent, structured data, and machine-readable context—ensuring your catalog is discoverable in AI-driven discovery channels such as ChatGPT, Google SGE, and other generative shopping experiences. - [How to Setup](https://docs.geoly.ai/shopify/how-to-setup-catalog-optimization.md): The entire process of using AI to batch audit product classifications and generate list-style descriptions. Ensure that your product data is correctly written and verified in the Shopify backend Metafields. ### Account & organization - [Basic Management Function](https://docs.geoly.ai/platform/basic-management.md): Configure GEOly account, organization, brand, and MCP settings using the current Settings navigation and permission scope. ### Integrations - [GA4 and AI Traffic Operation Guide](https://docs.geoly.ai/integrations/google-analytics-help-docs.md) - [Cloudflare Integration and AI Crawler Traffic](https://docs.geoly.ai/integrations/cloudflare-integration.md) ### Concepts - [What is llms.txt?](https://docs.geoly.ai/get-started/what-is-ai-llms-txt.md): Learn what the llms.txt proposal does, how its format differs from robots.txt, and how to test it without assuming traffic or ranking gains. ## MCP Server ### MCP Server - [GEOly MCP User Guide](https://docs.geoly.ai/developers/mcp.md): Connect GEOly MCP through OAuth to query authorized brand monitoring, citations, audits, free discovery, and paid public industry intelligence with transparent MCP Credits. ### CLI - [GEOly CLI: Use GEOly from Your Own Agent](https://docs.geoly.ai/developers/cli.md): Install GEOly CLI and authorize access for your agent. Protect locally stored credentials and receipts; hosted runs use AI Credits while billable data-tool calls use MCP Credits. ## API Reference ### Get started - [GEOly Data API](https://docs.geoly.ai/developers/data-api.md): Read monitoring data, synchronize completed answers, and create or run prompts through versioned REST resources with brand-bound API keys. ### Account - [Get brand](https://docs.geoly.ai/api-reference/account/get-brand.md): Brand profile, owning organization (plan, subscription status), AI platforms with completed answers in the last 30 days (entitled = covered by the plan), the first / last record day and first-collection progress. Use it to check which brand a key reads. Record days are UTC record-day keys (*_utc_dat… - [Get quota](https://docs.geoly.ai/api-reference/account/get-quota.md): MCP credits (the pool Data API calls are billed against) and AI credits of the organization that owns this API key: limit, used, remaining and reset time. Free; no query parameters. ### Catalog - [List topics](https://docs.geoly.ai/api-reference/catalog/list-topics.md): Every monitored topic of the brand with its active and archived prompt counts — the left rail of the in-app /prompts page. A small set, returned whole (no pagination). Prompts without a topic are not a row here: count them with GET /api/v1/prompts?topic_id=_none (page.total). Free; no query paramete… - [List prompts](https://docs.geoly.ai/api-reference/catalog/list-prompts.md): The prompt list straight from the prompt table: text, country, status, topic, tags, timestamps and whether the text names the brand — no metrics (use POST /api/v1/reports/prompts for per-prompt visibility). Ordered by (created_at, id) ascending with a true keyset cursor: a row whose (created_at, id)… - [Get prompt](https://docs.geoly.ai/api-reference/catalog/get-prompt.md): A single prompt by id, active or archived — the same row shape as GET /api/v1/prompts. 404 NOT_FOUND when the id does not exist for this key's brand. Free (no credits); no metrics are computed. Guarded at 200 requests per minute per organization: one bucket for GET /api/v1/prompts, GET /api/v1/promp… - [Create prompt](https://docs.geoly.ai/api-reference/catalog/create-prompt.md): Creates a prompt under the API key brand. Returns 201 and Location. Optional Idempotency-Key prevents duplicate execution; successful results replay for 24 hours. Unknown write outcomes are never automatically retried. tags_state reports whether every requested tag was read back (saved), some were n… ### Records - [List prompt records](https://docs.geoly.ai/api-reference/records/list-prompt-records.md): Paginated records, including pending and failed runs. At most 10,000 positions per query are reachable (page.capped); narrow the date window to continue. Cursors bind the brand, prompt, filters and limit. Pagination is not a snapshot: concurrent changes may skip or repeat records. Use explicit dates… - [Get latest records](https://docs.geoly.ai/api-reference/records/get-latest-records.md): One latest record per platform, without a date filter. A valid prompt with no records returns an empty array. Billed per call at the price in x-geoly-cost, using the existing MCP credit meter. Billing outcomes are in meta.quota. - [Get record](https://docs.geoly.ai/api-reference/records/get-record.md): Answer text is only returned when include=answer_text. Citations, search sources and shopping are bounded (30/30/20), with total counts and truncation flags. Internal provider errors and raw analysis are excluded. Responses exceeding 4 MB are rejected before settlement; omit include to fetch metadat… - [Sync records](https://docs.geoly.ai/api-reference/records/sync-records.md): Completed records ordered by changed_at then id, not by record date, so a late completion appears at its completion update. Uses a dedicated index and a safe watermark at least five minutes behind database time; a long-running database transaction that has already written holds the watermark at five… - [Trigger run](https://docs.geoly.ai/api-reference/records/trigger-run.md): Returns 202 when work was dispatched, otherwise 200 when naturally deduplicated. Location points to monitoring records. Consumes monitoring credits through the existing trigger. Refusals decided before anything is written (plan prompt limit, archived prompt, no allowed or available platform, expired… ### Reports - [Overview report](https://docs.geoly.ai/api-reference/reports/overview.md): Visibility score, mention rate, citation rate, share of voice and the per-platform breakdown for the window — the same numbers as the in-app overview. Every number counts the brands you track (your own brand + the competitors you added) under the current configuration; rank is among tracked brands o… - [Analytics report](https://docs.geoly.ai/api-reference/reports/analytics.md): Fixed columns per day (completed / mentioned answers, mention and citation rates, visibility score, sentiment buckets), optionally with the previous equal-length window. Rows carry both utc_date and business_date. Very long windows with breakdown=platform or topic may be capped (data.capped is then… - [Brand board](https://docs.geoly.ai/api-reference/reports/brand-board.md): Your brand and the competitors you track, ranked by mention rate in the window (caliber "tracked_brand_v1"; untracked brands never appear, rank is among tracked brands only). data.mode is "confirmed" for every filter combination ("scoped_open_world" is no longer returned); data.status says whether t… - [Citations report](https://docs.geoly.ai/api-reference/reports/citations.md): Top cited root domains with movers, your share, citation type split and the daily trend over closed business days. With a topic filter the whole-brand stats block is null and data.filtered is true. GET with body payload — no resources created. Billed per call at the price in x-geoly-cost (the same a… - [Prompts report](https://docs.geoly.ai/api-reference/reports/prompts.md): The in-app prompts table: per-prompt visibility, mention rate, citation rate, latest record per platform and window-over-window deltas, plus a summary of the whole filtered set. Paginate with cursor until page.next_cursor is null; the same filters and limit must be sent with every page. GET with bod… - [Prompt overview](https://docs.geoly.ai/api-reference/reports/prompt-overview.md): caliber and kpi_state distinguish legacy, mention-based and unavailable KPIs; with kpi_state own_pending the brand mention counts, rates and rank are null rather than placeholder zeros. Rates use percentages without conversion. Custom windows over 90 days are clamped by the tool; window echoes the a… ## OpenAPI Specs - [openapi](/api-reference/openapi.json) > The links below point to documentation indexes. Follow each `/_llms/` index recursively until you reach documentation pages. ## Indexes - [Chinese (40 pages)](https://docs.geoly.ai/_llms/zh.txt): Documentation for Chinese. This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.