{ "schema_version": 1, "name": "AI Data Connectors", "url": "https://ai-data-connector.com/", "publisher": { "name": "Olexander Paladiy", "type": "Person" }, "documentation": "https://ai-data-connector.com/about/", "ai_tools": { "evidence": [ { "id": "cpl-chatgpt", "url": "https://docs.coupler.io/destinations/categories/ai/chatgpt", "title": "ChatGPT destination", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-chatgpt-faq", "url": "https://docs.coupler.io/destinations/categories/ai/chatgpt/faq", "title": "ChatGPT destination FAQ", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-claude", "url": "https://docs.coupler.io/destinations/categories/ai/claude", "title": "Claude destination", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-claude-faq", "url": "https://docs.coupler.io/destinations/categories/ai/claude/faq", "title": "Claude destination FAQ", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-claude-free-pro", "url": "https://docs.coupler.io/troubleshooting/claude-connector-tools-dont-load-on-free-and-pro-plans", "title": "Claude connector tools do not load on Free and Pro plans", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-claude-issues", "url": "https://docs.coupler.io/destinations/categories/ai/claude/common-issues", "title": "Claude destination common issues", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-copilot-studio", "url": "https://docs.coupler.io/destinations/categories/ai/ms_copilot_studio", "title": "Microsoft Copilot Studio destination", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-cursor", "url": "https://docs.coupler.io/destinations/categories/ai/cursor", "title": "Cursor destination", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-custom-mcp", "url": "https://docs.coupler.io/destinations/categories/ai/custom_mcp", "title": "Custom MCP destination", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-custom-mcp-issues", "url": "https://docs.coupler.io/destinations/categories/ai/custom_mcp/common-issues", "title": "Custom MCP destination common issues", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-gemini-cli", "url": "https://docs.coupler.io/destinations/categories/ai/gemini", "title": "Gemini CLI destination", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-gemini-enterprise", "url": "https://docs.coupler.io/destinations/categories/ai/gemini_enterprise", "title": "Gemini Enterprise destination", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-mcp", "url": "https://docs.coupler.io/ai/mcp", "title": "What the Coupler.io MCP server can and cannot do", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-openclaw", "url": "https://docs.coupler.io/destinations/categories/ai/openclaw", "title": "OpenClaw destination", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-perplexity", "url": "https://docs.coupler.io/destinations/categories/ai/perplexity", "title": "Perplexity destination", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "dir-coupler-io", "url": "https://claude.com/connectors/coupler-io", "title": "Coupler.io in the Claude connectors directory", "publisher": "Anthropic", "checked_on": "2026-10-04", "kind": "directory" } ], "shared_notes": [ { "text": "Each AI tool sees only the datasets from data flows that have that tool as a destination. To use the same data in several tools, add each one as a destination on the data flow.", "evidence_ids": [ "cpl-mcp" ] }, { "text": "Every AI tool destination replaces the data on each run; append mode is not supported.", "evidence_ids": [ "cpl-claude", "cpl-chatgpt", "cpl-gemini-cli", "cpl-gemini-enterprise", "cpl-copilot-studio", "cpl-perplexity", "cpl-cursor", "cpl-openclaw", "cpl-custom-mcp" ] } ], "tools": [ { "id": "claude", "name": "Claude", "vendor": "Anthropic", "connection": "remote_mcp", "connection_label": "Remote MCP", "available_in": [ "Claude web", "Claude desktop", "Claude mobile", "Cowork", "Claude Code" ], "prerequisites": [ { "text": "A Claude account.", "evidence_ids": [ "cpl-claude" ] }, { "text": "On a Claude Team or Enterprise plan, admin access to add a connector.", "evidence_ids": [ "cpl-claude", "cpl-claude-issues" ] } ], "setups": [ { "title": "Claude web, desktop, and Cowork", "steps": [ { "text": "Install the Coupler.io connector from Claude's connector directory and click Connect.", "evidence_ids": [ "cpl-claude", "dir-coupler-io" ] }, { "text": "In the Connectors menu, click Connect again and sign in to Coupler.io.", "evidence_ids": [ "cpl-claude" ] }, { "text": "Open a new chat. Claude asks permission to use the Coupler.io tools the first time; allow them.", "evidence_ids": [ "cpl-claude", "cpl-claude-faq" ] } ] }, { "title": "Claude Code", "steps": [ { "text": "Start a new Claude Code session.", "evidence_ids": [ "cpl-claude" ] }, { "text": "Paste the connection command shown in the Claude destination step of your data flow.", "evidence_ids": [ "cpl-claude" ] }, { "text": "Restart the session.", "evidence_ids": [ "cpl-claude" ] }, { "text": "Run `/mcp`, select Coupler.io, and click Authorize in the browser window that opens.", "evidence_ids": [ "cpl-claude" ] } ] } ], "notes": [ { "text": "On individual Claude Free and Pro plans the official connector may connect but fail to load its tools. The documented workaround is a custom connector: in Coupler.io, open AI integrations, click Custom connector, and copy the URL. Remove the official connector in Claude, then go to Connectors, Add, Add custom connector, give it a name with plain letters and spaces only (for example Coupler Custom), paste the URL, and connect.", "evidence_ids": [ "cpl-claude-free-pro" ] }, { "text": "Local use with Claude desktop is also available through a Desktop Extension.", "evidence_ids": [ "cpl-claude" ] } ], "links": { "setup": "https://docs.coupler.io/destinations/categories/ai/claude", "directory": { "name": "the Claude connectors directory", "url": "https://claude.com/connectors/coupler-io" } } }, { "id": "chatgpt", "name": "ChatGPT", "vendor": "OpenAI", "connection": "remote_mcp", "connection_label": "Remote MCP", "connection_note": "Connects through the Coupler.io ChatGPT app.", "available_in": [], "prerequisites": [ { "text": "A ChatGPT account. Coupler.io says the app works on all ChatGPT plans, though the Free plan may not be enough for thorough analysis.", "evidence_ids": [ "cpl-chatgpt", "cpl-chatgpt-faq" ] } ], "setups": [ { "steps": [ { "text": "Open the Coupler.io app in ChatGPT, from the link in your Coupler.io account or the apps section of ChatGPT settings.", "evidence_ids": [ "cpl-chatgpt" ] }, { "text": "Install the app and authorize it with your Coupler.io account.", "evidence_ids": [ "cpl-chatgpt" ] }, { "text": "Start a new chat to pick up fresh data after a run, or ask ChatGPT to fetch it again in an ongoing conversation.", "evidence_ids": [ "cpl-chatgpt-faq" ] } ] } ], "notes": [], "links": { "setup": "https://docs.coupler.io/destinations/categories/ai/chatgpt" } }, { "id": "gemini-cli", "name": "Gemini CLI", "vendor": "Google", "connection": "remote_mcp", "connection_label": "Remote MCP", "available_in": [], "prerequisites": [ { "text": "Gemini CLI installed on your machine.", "evidence_ids": [ "cpl-gemini-cli" ] }, { "text": "A Google account with access to Gemini.", "evidence_ids": [ "cpl-gemini-cli" ] } ], "setups": [ { "steps": [ { "text": "In the Gemini CLI destination step, generate the command that adds the Coupler.io MCP server, then run it in your terminal. It looks like `gemini mcp add coupler --transport=http --scope=user https://mcp.coupler.io/mcp/...`.", "evidence_ids": [ "cpl-gemini-cli" ] }, { "text": "Start Gemini CLI and run `/mcp auth coupler` to sign in to Coupler.io.", "evidence_ids": [ "cpl-gemini-cli" ] } ] } ], "notes": [], "links": { "setup": "https://docs.coupler.io/destinations/categories/ai/gemini" } }, { "id": "gemini-enterprise", "name": "Gemini Enterprise", "vendor": "Google Cloud", "connection": "remote_mcp", "connection_label": "Remote MCP", "connection_note": "Added as a custom MCP data store with OAuth credentials from Coupler.io.", "available_in": [], "prerequisites": [ { "text": "A Google Cloud account with access to Gemini Enterprise.", "evidence_ids": [ "cpl-gemini-enterprise" ] }, { "text": "Admin or owner access in Coupler.io, to create the OAuth credentials.", "evidence_ids": [ "cpl-gemini-enterprise" ] } ], "setups": [ { "steps": [ { "text": "In the Gemini Enterprise destination step, create OAuth credentials and copy the Client ID and Client Secret. The secret is shown only once; if you lose it, regenerate it.", "evidence_ids": [ "cpl-gemini-enterprise" ] }, { "text": "In Gemini Enterprise, go to Data stores, Create data store, and choose Custom MCP server.", "evidence_ids": [ "cpl-gemini-enterprise" ] }, { "text": "Paste the MCP Server URL, Authorization URL, Token URL, and the scope `mcp` from the destination step, together with the Client ID and Client Secret.", "evidence_ids": [ "cpl-gemini-enterprise" ] } ] } ], "notes": [], "links": { "setup": "https://docs.coupler.io/destinations/categories/ai/gemini_enterprise", "vendor": "https://docs.cloud.google.com/gemini/enterprise/docs/connectors/custom-mcp-server/set-up-custom-mcp-server" } }, { "id": "copilot-studio", "name": "Microsoft Copilot Studio", "in_text": "your Copilot Studio agent", "vendor": "Microsoft", "connection": "remote_mcp", "connection_label": "Remote MCP", "connection_note": "Added to a Copilot Studio agent as an MCP tool.", "available_in": [], "prerequisites": [ { "text": "A Microsoft Copilot Studio environment with an agent you want to extend.", "evidence_ids": [ "cpl-copilot-studio" ] }, { "text": "Generative orchestration turned on for that agent, which is the default for new agents.", "evidence_ids": [ "cpl-copilot-studio" ] } ], "setups": [ { "steps": [ { "text": "Open your agent in Copilot Studio and confirm generative orchestration is on.", "evidence_ids": [ "cpl-copilot-studio" ] }, { "text": "Go to Tools, Add a tool, New tool, Model Context Protocol, and paste the Coupler.io MCP server URL shown in the destination step.", "evidence_ids": [ "cpl-copilot-studio" ] }, { "text": "For authentication, choose OAuth 2.0 and Dynamic discovery, then click Create. No client ID or secret is needed.", "evidence_ids": [ "cpl-copilot-studio" ] }, { "text": "The first time someone uses the tool, they see a consent card in chat and sign in to Coupler.io once.", "evidence_ids": [ "cpl-copilot-studio" ] } ] } ], "notes": [], "links": { "setup": "https://docs.coupler.io/destinations/categories/ai/ms_copilot_studio", "vendor": "https://learn.microsoft.com/en-us/microsoft-copilot-studio/mcp-add-existing-server-to-agent" } }, { "id": "perplexity", "name": "Perplexity", "connection": "local_mcp", "connection_label": "Local MCP", "connection_note": "Perplexity currently uses the local Coupler.io MCP server, which runs in Docker.", "available_in": [], "prerequisites": [ { "text": "A Perplexity account.", "evidence_ids": [ "cpl-perplexity" ] }, { "text": "Docker installed and running.", "evidence_ids": [ "cpl-perplexity" ] } ], "setups": [ { "steps": [ { "text": "In Perplexity, go to Settings, Connectors, click Add new connector, and choose the simpler settings.", "evidence_ids": [ "cpl-perplexity" ] }, { "text": "Name the connection coupler-io and enter the command `docker run --pull=always -e COUPLER_ACCESS_TOKEN --rm -i ghcr.io/railsware/coupler-io-mcp-server`.", "evidence_ids": [ "cpl-perplexity" ] }, { "text": "Add an environment variable called `COUPLER_ACCESS_TOKEN`, then generate a Coupler.io personal access token and paste it as the value.", "evidence_ids": [ "cpl-perplexity" ] }, { "text": "Save the configuration and accept the security warning about Docker.", "evidence_ids": [ "cpl-perplexity" ] } ] } ], "notes": [], "links": { "setup": "https://docs.coupler.io/destinations/categories/ai/perplexity" } }, { "id": "cursor", "name": "Cursor", "connection": "remote_mcp", "connection_label": "Remote MCP", "available_in": [], "prerequisites": [ { "text": "A Cursor account and the Cursor desktop app installed.", "evidence_ids": [ "cpl-cursor" ] } ], "setups": [ { "steps": [ { "text": "Open the Coupler.io entry in Cursor's MCP directory and click Add to Cursor.", "evidence_ids": [ "cpl-cursor" ] }, { "text": "In Cursor, find the new Coupler.io integration, click Needs authentication, then sign in with your Coupler.io account and grant access.", "evidence_ids": [ "cpl-cursor" ] }, { "text": "Open the AI pane and ask about your dataset.", "evidence_ids": [ "cpl-cursor" ] } ] } ], "notes": [], "links": { "setup": "https://docs.coupler.io/destinations/categories/ai/cursor", "directory": { "name": "the Cursor MCP directory", "url": "https://l.rw.rw/couplerio-cursor-mcp" } } }, { "id": "openclaw", "name": "OpenClaw", "connection": "remote_mcp", "connection_label": "Remote MCP", "connection_note": "Connects to the Coupler.io MCP server through mcporter.", "available_in": [], "prerequisites": [ { "text": "OpenClaw installed on your machine.", "evidence_ids": [ "cpl-openclaw" ] } ], "setups": [ { "steps": [ { "text": "Install the coupler-io skill from ClawHub. It holds the commands that connect the Coupler.io MCP server to OpenClaw with mcporter.", "evidence_ids": [ "cpl-openclaw" ] } ] } ], "notes": [], "links": { "setup": "https://docs.coupler.io/destinations/categories/ai/openclaw", "directory": { "name": "ClawHub", "url": "https://clawhub.ai/nika-is-nika/coupler-io" } } }, { "id": "custom-mcp", "name": "Custom MCP", "in_text": "your MCP client", "connection": "remote_mcp", "connection_label": "Remote MCP", "connection_note": "Connects any MCP-compatible client with a personal access token.", "available_in": [], "prerequisites": [ { "text": "An MCP-compatible client, such as Claude, Cursor, Zapier, n8n, Make, or your own application.", "evidence_ids": [ "cpl-custom-mcp" ] }, { "text": "Basic understanding of how to configure that client to connect to an external server.", "evidence_ids": [ "cpl-custom-mcp" ] } ], "setups": [ { "title": "your MCP client", "steps": [ { "text": "In the Custom MCP destination step, click Generate token and copy it together with the MCP server URL.", "evidence_ids": [ "cpl-custom-mcp" ] }, { "text": "Add the token and URL to your MCP client's JSON config, following the example config shown there.", "evidence_ids": [ "cpl-custom-mcp" ] } ] } ], "notes": [ { "text": "Treat the personal access token as a secret and do not share it.", "evidence_ids": [ "cpl-custom-mcp" ] }, { "text": "Behind a corporate firewall, allow outbound connections to the Coupler.io MCP endpoint.", "evidence_ids": [ "cpl-custom-mcp-issues" ] } ], "links": { "setup": "https://docs.coupler.io/destinations/categories/ai/custom_mcp" } } ] }, "sources": [ { "id": "facebook-ads", "slug": "facebook-ads", "name": "Facebook Ads", "aliases": [ "Meta Ads", "Facebook Ads Manager" ], "url": "https://ai-data-connector.com/sources/facebook-ads/", "summary": "Facebook Ads (Meta Ads) is Meta's advertising platform for campaigns across Facebook, Instagram, Messenger, and the Audience Network. Its data covers campaigns, ad sets, ads, creatives, spend, reach, clicks, and conversions. To analyze it in ChatGPT, Claude, Gemini, or another LLM, use scheduled Coupler.io imports.", "research": { "searched_on": "2026-10-04", "coverage_note": "One route is covered, researched on 4 October 2026: the Coupler.io data platform. No dedicated Facebook Ads or Meta Ads entry was reachable in Claude's connector directory at the obvious addresses; the directory cannot be searched exhaustively, so no claim is made that none exists." }, "evidence": [ { "id": "cpl-claude-destination", "url": "https://docs.coupler.io/destinations/categories/ai/claude", "title": "Claude destination", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-claude-faq", "url": "https://docs.coupler.io/destinations/categories/ai/claude/faq", "title": "Claude destination FAQ", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-claude-issues", "url": "https://docs.coupler.io/destinations/categories/ai/claude/common-issues", "title": "Claude destination common issues", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-fb-best", "url": "https://docs.coupler.io/sources/category/ppc/facebook-ads/best-practices", "title": "Facebook Ads best practices", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-fb-claude-page", "url": "https://www.coupler.io/claude-integrations/facebook-ads-to-claude", "title": "Connect Facebook Ads to Claude", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-fb-data", "url": "https://docs.coupler.io/sources/category/ppc/facebook-ads/data-overview", "title": "Facebook Ads data overview", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-fb-faq", "url": "https://docs.coupler.io/sources/category/ppc/facebook-ads/faq", "title": "Facebook Ads source FAQ", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-fb-issues", "url": "https://docs.coupler.io/sources/category/ppc/facebook-ads/common-issues", "title": "Facebook Ads source common issues", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-fb-overview", "url": "https://docs.coupler.io/sources/category/ppc/facebook-ads", "title": "Facebook Ads source overview", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-mcp", "url": "https://docs.coupler.io/ai/mcp", "title": "What the Coupler.io MCP server can and cannot do", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-refresh", "url": "https://docs.coupler.io/functionality/flow-settings/how-to-set-up-automatic-data-refresh", "title": "How to set up automatic data refresh", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "dir-coupler-io", "url": "https://claude.com/connectors/coupler-io", "title": "Coupler.io in the Claude connectors directory", "publisher": "Anthropic", "checked_on": "2026-10-04", "kind": "directory" } ], "options": [ { "id": "coupler-io", "name": "Coupler.io", "provider": "Coupler.io", "method": "data_platform", "method_label": "Data platform", "maintainer": { "status": "known", "value": "third_party", "evidence_ids": [ "cpl-mcp", "cpl-claude-destination" ], "note": "Coupler.io builds and runs this connector and its MCP server." }, "maintainer_label": "Third-party", "route_status": { "status": "known", "value": "available", "evidence_ids": [ "cpl-claude-destination", "cpl-fb-claude-page" ] }, "directory_listing": { "status": "known", "value": "https://claude.com/connectors/coupler-io", "evidence_ids": [ "dir-coupler-io" ], "note": "The entry is marked Anthropic verified, credited to Coupler.io, added in November 2025, and records sign-in as required. It lists four tools: get-data, get-schema, list-dataflows, and get-dataflow" }, "links": { "overview": "https://docs.coupler.io/sources/category/ppc/facebook-ads" }, "works_with": [ "claude", "chatgpt", "gemini-cli", "gemini-enterprise", "copilot-studio", "perplexity", "cursor", "openclaw", "custom-mcp" ], "access": { "status": "known", "value": "read", "evidence_ids": [ "cpl-mcp", "dir-coupler-io" ], "note": "Queries against imported data are read-only, and nothing in either source describes changing campaigns, budgets, or bids. Coupler.io's own documentation also describes tools that change your Coupler.io workspace, such as creating a data flow or triggering a run" }, "data_available": { "status": "known", "value": [ "Nine report types: Reports and Insights, List of Sponsored Leads, List of Campaigns, List of Ad Sets, List of Ads, List of Ad Creatives, List of Ads with Ad Creatives, List of Activities, and List of Business invoices", "Reports and Insights carries the performance data: account, campaign, ad set, and ad dimensions with performance, click, cost, engagement, video, and conversion metrics", "Conversion events come in Total, Unique, Cost, Unique Cost, and Value variants, with purchase revenue as the Value variant and Purchase ROAS as its own metric; custom conversions and custom events are included", "Breakdowns: age, gender, country, region, Comscore markets, publisher platform, impression device, device platform, product id, frequency value, and hourly stats", "Time splits: daily, weekly, monthly, or totals only for the whole date range", "Structural lists cover campaign, ad set, and ad statuses, budgets, objectives, and targeting; creatives carry image URLs, ad text, and UTM url_tags", "List of Activities is the account audit log, including billing and funding activity; List of Business invoices gives monthly invoices with billed amounts per campaign" ], "evidence_ids": [ "cpl-fb-overview", "cpl-fb-data", "cpl-fb-faq" ], "note": "The structural and creative lists carry no performance metrics, so creative performance needs a Reports and Insights import at the Ad level joined to the creative list on Ad Id" }, "history": { "status": "known", "value": [ "You set a start and end date; no hard limit on the range is stated, but very large ranges with many breakdowns can be slow or hit API limits", "For initial backfills the documentation recommends pulling data in monthly chunks" ], "evidence_ids": [ "cpl-fb-faq", "cpl-fb-best" ], "note": "no fixed Facebook Ads retention window is stated" }, "refresh": { "status": "known", "value": [ "A successful manual run is required before a schedule can be set", "Scheduled refresh is described as hourly, daily, or a custom interval", "The platform supports intervals from every 15 minutes to monthly, depending on plan" ], "evidence_ids": [ "cpl-claude-faq", "cpl-refresh" ], "note": "Which intervals each plan allows is not documented" }, "multi_source": { "status": "known", "value": [ "One data flow can take several sources and combine them with Join or Append before the data reaches Claude", "Facebook Ads is documented as merging with Google Ads, Google Analytics, CRM, or e-commerce data for a full-funnel view of marketing", "Creative details and performance come from separate report types and are joined on Ad Id" ], "evidence_ids": [ "cpl-claude-faq", "cpl-fb-overview", "cpl-fb-faq" ] }, "prerequisites": { "status": "known", "value": [ "A Facebook account with access to at least one ad account in Meta Business Suite", "Admin or Advertiser permissions on the ad account you want to pull data from", "A Coupler.io account" ], "evidence_ids": [ "cpl-fb-overview", "cpl-fb-issues", "cpl-claude-destination" ] }, "pricing": { "status": "unknown", "value": null, "evidence_ids": [], "note": "no pricing page was checked for this record" }, "data_path": { "status": "known", "value": "Coupler.io reads the Facebook Ads API, stores the result as a dataset, and serves it to Claude over its MCP server. Queries run on Coupler.io's side, so a large dataset does not have to fit into the model's context. Claude sees only datasets from data flows that have Claude as their destination.", "evidence_ids": [ "cpl-mcp" ] }, "limits": { "status": "known", "value": [ "The Facebook Ads API has a 1 to 2 day data latency, so today and sometimes yesterday are incomplete", "Attribution windows must match Ads Manager for the numbers to agree; if none is selected in Coupler.io, a 7-day click window is used", "The 1-day engaged-view attribution window is not available through the API, so conversion counts can be lower than in Ads Manager", "Reach and ratio metrics such as CTR, CPC, and CPM cannot be summed across rows", "Since 6 August 2026, Meta returns no data for the Impression device, Frequency value, and Hourly stats by audience time zone breakdowns until they are enabled per ad account in Ads Manager", "Demographic or geographic breakdowns cannot be combined with hourly breakdowns, and action breakdowns can make the parts add up to more than the total", "Any filter in Advanced settings makes the API return active objects only, unless an explicit status filter is added", "Several data flows pulling from one ad account at once can hit Facebook's per-account rate limits; large reports can fail with Async Job failed", "Currency follows the ad account and cannot be converted, and insights are returned in the ad account's timezone" ], "evidence_ids": [ "cpl-fb-data", "cpl-fb-faq", "cpl-fb-issues", "cpl-fb-best" ] }, "capabilities": [], "setup_steps": [ { "text": "In Coupler.io, create a data flow, add Facebook Ads as the source, and sign in with your Facebook account, granting every permission it requests.", "evidence_ids": [ "cpl-fb-overview", "cpl-fb-issues" ] }, { "text": "Select the ad accounts to pull from, then choose a report type; Reports and Insights is the one with performance metrics.", "evidence_ids": [ "cpl-fb-overview", "cpl-fb-best" ] }, { "text": "Set the date range with yesterday as the end date, pick your metrics and breakdowns, and match the attribution window to the one you use in Ads Manager.", "evidence_ids": [ "cpl-fb-overview", "cpl-fb-best" ] }, { "text": "Choose {destination} as the destination in the data flow's destination step.", "evidence_ids": [ "cpl-claude-destination" ] }, { "text": "Give the data flow a clear name, because {tool} refers to the dataset by that name.", "evidence_ids": [ "cpl-claude-destination" ] }, { "text": "Run the data flow manually once and wait for it to finish successfully.", "evidence_ids": [ "cpl-claude-destination" ] }, { "text": "Set a refresh schedule once that first run has succeeded.", "evidence_ids": [ "cpl-claude-faq" ] } ], "success_check": { "text": "Ask {tool} to list the datasets it can see; the Facebook Ads data flow should appear under the name you gave it. Then compare total spend for a closed date range with Ads Manager, using the same attribution window.", "evidence_ids": [ "cpl-claude-destination", "cpl-fb-faq" ] } } ], "recommendations": [ { "job": "Reporting on paid social spend over time or alongside Google Ads, analytics, and CRM data", "option_id": "coupler-io", "reason": "Queries run against a stored dataset rather than the Facebook Ads API, so long date ranges do not have to fit into the conversation, and one data flow can join Facebook Ads with Google Ads, Google Analytics, or CRM data before Claude sees it.", "evidence_ids": [ "cpl-mcp", "cpl-fb-overview" ] } ], "faq": [ { "question": "Can Claude pause an ad or change my budgets?", "answer": "No. Coupler.io is read-only against Facebook Ads: Claude queries an imported dataset and cannot change campaigns, budgets, or bids.", "evidence_ids": [ "cpl-mcp" ] }, { "question": "Why don't my numbers match Ads Manager?", "answer": "Usually the attribution window. Coupler.io uses a 7-day click window when none is selected, so set the same window you use in Ads Manager. Then check that the period split, the entity level, and the breakdowns match. Conversions attributed to the 1-day engaged-view window are not available through the API at all, so those counts can stay slightly lower.", "evidence_ids": [ "cpl-fb-issues", "cpl-fb-faq" ] }, { "question": "Why is today's data missing or incomplete?", "answer": "The Facebook Ads API has a 1 to 2 day data latency, so the current day and sometimes yesterday are not fully processed. Set the end date to yesterday for complete numbers.", "evidence_ids": [ "cpl-fb-faq", "cpl-fb-best" ] }, { "question": "Why is my Reach total higher than in Ads Manager?", "answer": "Reach counts unique people, so adding up daily Reach rows counts anyone who saw an ad on several days more than once. Use the Totals only period split for a single deduplicated figure over a fixed date range, or keep Reach per row. Ratio metrics such as CTR, CPC, and CPM should likewise be recalculated from totals rather than summed.", "evidence_ids": [ "cpl-fb-faq", "cpl-fb-issues" ] }, { "question": "Why did the Impression device, Frequency value, or Hourly stats breakdown stop returning data?", "answer": "Since 6 August 2026, Meta requires these breakdowns to be enabled per ad account in Ads Manager, and until then its API returns no results for them, even if they worked before. Enable the breakdown for the ad account and re-run the data flow; full history, including data from before enablement, comes back.", "evidence_ids": [ "cpl-fb-issues", "cpl-fb-faq" ] }, { "question": "How are ad accounts counted toward my Coupler.io plan?", "answer": "Facebook Ads is a tenant-based source, so each ad account you select counts as one account toward your plan limit, however many report types or data flows use it.", "evidence_ids": [ "cpl-fb-faq" ] } ], "related_skills": [], "related_source_ids": [] }, { "id": "google-ads", "slug": "google-ads", "name": "Google Ads", "aliases": [ "AdWords", "Google Adwords" ], "url": "https://ai-data-connector.com/sources/google-ads/", "summary": "Google Ads is Google's advertising platform for search, display, and video campaigns. Its data covers campaigns, ad groups, ads, keywords, spend, clicks, impressions, and conversions. To analyze it in ChatGPT, Claude, Gemini, or another LLM, use scheduled Coupler.io imports.", "research": { "searched_on": "2026-10-04", "coverage_note": "One route is covered, researched on 4 October 2026: the Coupler.io data platform. No dedicated Google Ads entry was reachable in Claude's connector directory at the obvious address; the directory cannot be searched exhaustively, so no claim is made that none exists." }, "evidence": [ { "id": "cpl-ads-best", "url": "https://docs.coupler.io/sources/category/ppc/google-ads/best-practices", "title": "Google Ads best practices", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-ads-claude-page", "url": "https://www.coupler.io/claude-integrations/google-ads-to-claude", "title": "Connect Google Ads to Claude", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-ads-data", "url": "https://docs.coupler.io/sources/category/ppc/google-ads/data-overview", "title": "Google Ads data overview", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-ads-faq", "url": "https://docs.coupler.io/sources/category/ppc/google-ads/faq", "title": "Google Ads source FAQ", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-ads-issues", "url": "https://docs.coupler.io/sources/category/ppc/google-ads/common-issues", "title": "Google Ads source common issues", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-ads-overview", "url": "https://docs.coupler.io/sources/category/ppc/google-ads", "title": "Google Ads source overview", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-claude-destination", "url": "https://docs.coupler.io/destinations/categories/ai/claude", "title": "Claude destination", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-claude-faq", "url": "https://docs.coupler.io/destinations/categories/ai/claude/faq", "title": "Claude destination FAQ", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-claude-issues", "url": "https://docs.coupler.io/destinations/categories/ai/claude/common-issues", "title": "Claude destination common issues", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-mcp", "url": "https://docs.coupler.io/ai/mcp", "title": "What the Coupler.io MCP server can and cannot do", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-refresh", "url": "https://docs.coupler.io/functionality/flow-settings/how-to-set-up-automatic-data-refresh", "title": "How to set up automatic data refresh", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "dir-coupler-io", "url": "https://claude.com/connectors/coupler-io", "title": "Coupler.io in the Claude connectors directory", "publisher": "Anthropic", "checked_on": "2026-10-04", "kind": "directory" } ], "options": [ { "id": "coupler-io", "name": "Coupler.io", "provider": "Coupler.io", "method": "data_platform", "method_label": "Data platform", "maintainer": { "status": "known", "value": "third_party", "evidence_ids": [ "cpl-mcp", "cpl-claude-destination" ], "note": "Coupler.io builds and runs this connector and its MCP server." }, "maintainer_label": "Third-party", "route_status": { "status": "known", "value": "available", "evidence_ids": [ "cpl-claude-destination", "cpl-ads-claude-page" ] }, "directory_listing": { "status": "known", "value": "https://claude.com/connectors/coupler-io", "evidence_ids": [ "dir-coupler-io" ], "note": "The entry is marked Anthropic verified, credited to Coupler.io, added in November 2025, and records sign-in as required. It lists four tools: get-data, get-schema, list-dataflows, and get-dataflow" }, "links": { "overview": "https://docs.coupler.io/sources/category/ppc/google-ads" }, "works_with": [ "claude", "chatgpt", "gemini-cli", "gemini-enterprise", "copilot-studio", "perplexity", "cursor", "openclaw", "custom-mcp" ], "access": { "status": "known", "value": "read", "evidence_ids": [ "cpl-mcp", "dir-coupler-io" ], "note": "Queries against imported data are read-only, and nothing in either source describes changing campaigns, budgets, or bids. Coupler.io's own documentation also describes tools that change your Coupler.io workspace, such as creating a data flow or triggering a run" }, "data_available": { "status": "known", "value": [ "Over 30 report types in total", "Frequently used: Campaign performance, Keywords performance, Ad group performance, Ad performance, Asset group performance, Custom GAQL", "Audience and demographics: Ad group audience performance, Campaign audience performance, Age range performance, Gender performance, User interest and user list", "Search and placement: Search query performance, Display keyword performance, Display topics performance, Placement performance, Landing page and Expanded landing page", "Geography and network: Geographic performance by country, state, or region, User location performance, Campaign performance by ad network type", "Other performance: Account performance, Click performance, Shopping performance, Responsive search ad performance, Video campaign performance", "Structural: Accounts, Campaigns, Ad groups, Ads. Settings: Campaign budgets, Campaign criterion, Campaign labels, Bidding strategies, Labels", "Metric categories: Budget, Clicks, Conversions, Cost, Performance, Cross sell and Sales, Video, Phone", "Time splits: Date, Day of week, Week, Month, Quarter, Year" ], "evidence_ids": [ "cpl-ads-data" ], "note": "Structural reports and Custom GAQL do not support time period splits, so one row per day is not available for them" }, "history": { "status": "known", "value": [ "You set a start and end date; no hard limit on the range is stated, but long ranges with daily splits are slow", "For initial backfills the documentation recommends pulling in monthly or quarterly chunks to avoid API limits" ], "evidence_ids": [ "cpl-ads-faq", "cpl-ads-best" ], "note": "no fixed Google Ads retention window is stated" }, "refresh": { "status": "known", "value": [ "A successful manual run is required before a schedule can be set", "Scheduled refresh is described as hourly, daily, or a custom interval", "The platform supports intervals from every 15 minutes to monthly, depending on plan" ], "evidence_ids": [ "cpl-claude-faq", "cpl-refresh" ], "note": "Which intervals each plan allows is not documented" }, "multi_source": { "status": "known", "value": [ "One data flow can take several sources and combine them with Join or Append before the data reaches Claude", "Google Ads is documented as merging with Google Analytics, Facebook Ads, CRM, or e-commerce data for a full-funnel view of spend and results" ], "evidence_ids": [ "cpl-claude-faq", "cpl-ads-overview" ] }, "prerequisites": { "status": "known", "value": [ "A Coupler.io account" ], "evidence_ids": [ "cpl-claude-destination" ] }, "pricing": { "status": "unknown", "value": null, "evidence_ids": [], "note": "no pricing page was checked for this record" }, "data_path": { "status": "known", "value": "Coupler.io reads the Google Ads API, stores the result as a dataset, and serves it to Claude over its MCP server. Queries run on Coupler.io's side, so a large dataset does not have to fit into the model's context. Claude sees only datasets from data flows that have Claude as their destination.", "evidence_ids": [ "cpl-mcp" ] }, "limits": { "status": "known", "value": [ "Google Ads API quota can be exceeded globally or per account, returning a 429 with a retry delay; the documented fix is to wait roughly 4 hours for the quota to reset", "Rows with all-zero metrics are not returned by the API, so an ad or keyword with no activity in the period simply does not appear", "Low-volume search terms are grouped under Other search terms to meet Google's privacy thresholds", "Custom columns created in the Google Ads interface are not part of the API and cannot be pulled", "Impression share metrics are not summable across rows", "Performance Max campaigns have asset groups rather than traditional ads and ad groups", "Custom GAQL returns cost as cost_micros, which must be divided by 1,000,000", "Data is typically available within a few hours, but conversions may take 24 to 72 hours to finalise" ], "evidence_ids": [ "cpl-ads-data", "cpl-ads-issues", "cpl-ads-faq" ] }, "capabilities": [], "setup_steps": [], "success_check": { "text": "Ask {tool} to list the datasets it can see. Allow the Coupler.io tools if it asks, and your Coupler.io datasets are listed.", "evidence_ids": [ "cpl-claude-faq", "cpl-claude-issues" ] } } ], "recommendations": [ { "job": "Reporting on spend over time or alongside analytics and CRM data", "option_id": "coupler-io", "reason": "Queries run against a stored dataset rather than the Ads API, which sidesteps the quota errors that bite on large backfills, and one data flow can join Google Ads with GA4 or CRM data before Claude sees it.", "evidence_ids": [ "cpl-mcp", "cpl-ads-issues", "cpl-ads-overview" ] } ], "faq": [ { "question": "Can Claude pause a campaign or change my bids?", "answer": "No. Coupler.io is read-only against Google Ads: Claude queries an imported dataset and cannot change campaigns, budgets, or bids.", "evidence_ids": [ "cpl-mcp" ] }, { "question": "Why does an import fail with a quota error?", "answer": "The Google Ads API quota can be exhausted globally or per account, returning a 429 with a retry delay. The documented fix is to wait roughly 4 hours for the quota to reset, and to pull large backfills in monthly or quarterly chunks rather than years at once.", "evidence_ids": [ "cpl-ads-issues", "cpl-ads-best" ] }, { "question": "Why are some campaigns or keywords missing from my data?", "answer": "The Google Ads API does not return rows where every metric is zero, so anything with no activity in the period simply does not appear. Separately, low-volume search terms are grouped under Other search terms to meet Google's privacy thresholds, and custom columns built in the Google Ads interface are not available through the API at all.", "evidence_ids": [ "cpl-ads-issues", "cpl-ads-data" ] }, { "question": "Why do my conversion numbers keep changing?", "answer": "Google Ads data is typically available within a few hours, but conversions may take 24 to 72 hours to finalise. A report pulled this morning about yesterday will often disagree with the same report pulled next week.", "evidence_ids": [ "cpl-ads-data", "cpl-ads-faq" ] }, { "question": "Does my manager account give Claude access to all my sub-accounts?", "answer": "No. Manager (MCC) access does not automatically grant data access to sub-accounts; each ad account needs explicit Standard or Admin access. Note also that each ad account you select counts as one account toward your Coupler.io plan limit, because Google Ads is a tenant-based source.", "evidence_ids": [ "cpl-ads-issues", "cpl-ads-faq" ] } ], "related_skills": [], "related_source_ids": [] }, { "id": "google-analytics-4", "slug": "google-analytics-4", "name": "Google Analytics 4", "aliases": [ "GA4", "Google Analytics" ], "url": "https://ai-data-connector.com/sources/google-analytics-4/", "summary": "Google Analytics 4 (GA4) is Google's web and app analytics product. Its data covers traffic, users, sessions, events, conversions, and acquisition sources. To analyze it in ChatGPT, Claude, Gemini, or another LLM, use scheduled Coupler.io imports.", "research": { "searched_on": "2026-10-04", "coverage_note": "One route is covered, researched on 4 October 2026: the Coupler.io data platform. No dedicated GA4 entry was reachable in Claude's connector directory at the obvious addresses; the directory cannot be searched exhaustively, so no claim is made that none exists." }, "evidence": [ { "id": "cpl-claude-destination", "url": "https://docs.coupler.io/destinations/categories/ai/claude", "title": "Claude destination", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-claude-faq", "url": "https://docs.coupler.io/destinations/categories/ai/claude/faq", "title": "Claude destination FAQ", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-claude-free-pro", "url": "https://docs.coupler.io/troubleshooting/claude-connector-tools-dont-load-on-free-and-pro-plans", "title": "Claude connector tools do not load on Free and Pro plans", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-claude-issues", "url": "https://docs.coupler.io/destinations/categories/ai/claude/common-issues", "title": "Claude destination common issues", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-ga4-claude-page", "url": "https://www.coupler.io/claude-integrations/google-analytics-to-claude", "title": "Connect Google Analytics 4 to Claude", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-ga4-data", "url": "https://docs.coupler.io/sources/category/ppc/google-analytics/data-overview", "title": "Google Analytics 4 data overview", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-ga4-faq", "url": "https://docs.coupler.io/sources/category/ppc/google-analytics/faq", "title": "Google Analytics 4 source FAQ", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-ga4-issues", "url": "https://docs.coupler.io/sources/category/ppc/google-analytics/common-issues", "title": "Google Analytics 4 source common issues", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-ga4-overview", "url": "https://docs.coupler.io/sources/category/ppc/google-analytics", "title": "Google Analytics 4 source overview", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-mcp", "url": "https://docs.coupler.io/ai/mcp", "title": "What the Coupler.io MCP server can and cannot do", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-refresh", "url": "https://docs.coupler.io/functionality/flow-settings/how-to-set-up-automatic-data-refresh", "title": "How to set up automatic data refresh", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "dir-coupler-io", "url": "https://claude.com/connectors/coupler-io", "title": "Coupler.io in the Claude connectors directory", "publisher": "Anthropic", "checked_on": "2026-10-04", "kind": "directory" } ], "options": [ { "id": "coupler-io", "name": "Coupler.io", "provider": "Coupler.io", "method": "data_platform", "method_label": "Data platform", "maintainer": { "status": "known", "value": "third_party", "evidence_ids": [ "cpl-mcp", "cpl-claude-destination" ], "note": "Coupler.io builds and runs this connector and its MCP server." }, "maintainer_label": "Third-party", "route_status": { "status": "known", "value": "available", "evidence_ids": [ "cpl-claude-destination", "cpl-ga4-claude-page" ] }, "directory_listing": { "status": "known", "value": "https://claude.com/connectors/coupler-io", "evidence_ids": [ "dir-coupler-io" ], "note": "The entry is marked Anthropic verified, credited to Coupler.io, added in November 2025, and records sign-in as required. It lists four tools: get-data, get-schema, list-dataflows, and get-dataflow" }, "links": { "overview": "https://docs.coupler.io/sources/category/ppc/google-analytics" }, "works_with": [ "claude", "chatgpt", "gemini-cli", "gemini-enterprise", "copilot-studio", "perplexity", "cursor", "openclaw", "custom-mcp" ], "access": { "status": "known", "value": "read", "evidence_ids": [ "cpl-mcp", "dir-coupler-io" ], "note": "Queries against imported data are read-only, and nothing in either source describes writing back to GA4. Coupler.io's own documentation also describes tools that change your Coupler.io workspace, such as creating a data flow or triggering a run" }, "data_available": { "status": "known", "value": [ "One flexible report type that you build from your own choice of metrics and dimensions, loaded dynamically from the GA4 API for your property", "Traffic metrics: Views, Sessions, Total users, New users, Active users, Bounce rate, Engagement rate, Average session duration, Views per session, Sessions per user, Event count, Key events", "Ecommerce metrics: Purchase revenue, Total revenue, Ecommerce purchases, Add-to-carts, Checkouts, Item revenue, Transactions, Average purchase revenue", "Advertising metrics: Ads cost, Ads clicks, Ads impressions, Return on ad spend, Cost per key event", "Dimensions: date and time splits, session source, medium, campaign and default channel group, country, city, region, device, browser, operating system, page path, page title, landing page, hostname, and item-level ecommerce fields" ], "evidence_ids": [ "cpl-ga4-data" ] }, "history": { "status": "known", "value": [ "GA4 properties have a configurable retention period of either 2 months or 14 months, and only data inside that window can be pulled", "You set a start and end date for the report; no hard limit on the range is stated, but long ranges with many dimensions are slow" ], "evidence_ids": [ "cpl-ga4-faq" ], "note": "Preserving data beyond GA4's retention window means exporting it before it expires, which is an argument for scheduling an import rather than querying on demand" }, "refresh": { "status": "known", "value": [ "A successful manual run is required before a schedule can be set", "Scheduled refresh is described as hourly, daily, or a custom interval", "The platform supports intervals from every 15 minutes to monthly, depending on plan" ], "evidence_ids": [ "cpl-claude-faq", "cpl-refresh" ], "note": "Which intervals each plan allows is not documented" }, "multi_source": { "status": "known", "value": [ "One data flow can take several sources and combine them with Join or Append before the data reaches Claude", "GA4 is documented as merging with Google Ads, Facebook Ads, CRM, or e-commerce data for a full-funnel view", "Running several reports and joining them is the documented workaround for GA4's 10-metric ceiling" ], "evidence_ids": [ "cpl-claude-faq", "cpl-ga4-overview", "cpl-ga4-data" ] }, "prerequisites": { "status": "known", "value": [ "A Google account with access to at least one GA4 property", "Viewer permissions or higher on the GA4 property", "The analytics.readonly permission must be granted during the OAuth flow", "A Coupler.io account" ], "evidence_ids": [ "cpl-ga4-overview", "cpl-ga4-issues", "cpl-claude-destination" ], "note": "no GA4 plan tier, such as standard versus 360, is named" }, "pricing": { "status": "unknown", "value": null, "evidence_ids": [], "note": "no pricing page was checked for this record" }, "data_path": { "status": "known", "value": "Coupler.io reads the GA4 Data API, stores the result as a dataset, and serves it to Claude over its MCP server. Queries run on Coupler.io's side, so a large dataset does not have to fit into the model's context. Claude sees only datasets from data flows that have Claude as their destination.", "evidence_ids": [ "cpl-mcp" ] }, "limits": { "status": "known", "value": [ "Up to 10 metrics and 9 dimensions per data flow, which is a GA4 API limit rather than a Coupler.io one", "Not all metric and dimension pairs are compatible; an invalid combination returns a 400 error", "The GA4 Data API applies request quotas per property, so several flows pulling the same property at once can hit them", "GA4 may withhold rows representing too few users through thresholding, especially when Google Signals is enabled", "Data is typically available within a few hours, but some metrics take 24 to 48 hours to finish processing", "Total users broken down by a dimension will not sum to the overall total", "On the free plan only the first 1,000 rows are exported, which shows up as gaps in dates" ], "evidence_ids": [ "cpl-ga4-data", "cpl-ga4-issues" ], "note": "Data is pulled through the GA4 Data API, which returns unsampled results" }, "capabilities": [], "setup_steps": [ { "text": "In Coupler.io, create a data flow and select Google Analytics 4 as the source.", "evidence_ids": [ "cpl-ga4-overview" ] }, { "text": "Choose your property, then pick up to 10 metrics and 9 dimensions and set the date range.", "evidence_ids": [ "cpl-ga4-data" ] }, { "text": "Choose {destination} as the destination in the data flow's destination step.", "evidence_ids": [ "cpl-claude-destination" ] }, { "text": "Give the data flow a clear name, because {tool} refers to the dataset by that name.", "evidence_ids": [ "cpl-claude-destination" ] }, { "text": "Run the data flow manually once and wait for it to finish successfully.", "evidence_ids": [ "cpl-claude-destination" ] }, { "text": "Set a refresh schedule once that first run has succeeded.", "evidence_ids": [ "cpl-claude-faq" ] } ], "success_check": { "text": "Ask {tool} to list the datasets it can see; the GA4 data flow should appear under the name you gave it. Then compare the row count {tool} reports with the run in Coupler.io, because the preview alone does not confirm what {tool} received.", "evidence_ids": [ "cpl-claude-destination", "cpl-claude-issues" ] } } ], "recommendations": [ { "job": "Tracking GA4 trends over time or alongside ad spend", "option_id": "coupler-io", "reason": "Scheduled imports keep a copy of data that GA4 itself deletes after 2 or 14 months, queries run against the stored dataset rather than the API, and one data flow can join GA4 with Google Ads or CRM data before Claude sees it.", "evidence_ids": [ "cpl-ga4-faq", "cpl-mcp", "cpl-ga4-overview" ] } ], "faq": [ { "question": "Can Claude change my GA4 configuration?", "answer": "No. Coupler.io is read-only against GA4: Claude queries an imported dataset and cannot edit your Analytics configuration, settings, or events.", "evidence_ids": [ "cpl-mcp" ] }, { "question": "Why can I only pick 10 metrics?", "answer": "That ceiling is the GA4 Data API's: up to 10 metrics and 9 dimensions per report. The documented workaround is to run more than one report and join them.", "evidence_ids": [ "cpl-ga4-data" ] }, { "question": "How far back can Claude see my analytics?", "answer": "As far as GA4 keeps it. A GA4 property retains data for either 2 or 14 months depending on its setting, and no route can read past that. If you need longer history, the documented approach is to export on a schedule and store the results before they expire.", "evidence_ids": [ "cpl-ga4-faq" ] }, { "question": "Will the numbers Claude reports match the GA4 interface?", "answer": "Usually, with two documented caveats. GA4 may withhold rows representing too few users through thresholding, especially when Google Signals is on, and Total users broken down by a dimension will not sum to the overall total. Data pulled through the Data API is unsampled, and most metrics settle within a few hours, though some take 24 to 48 hours.", "evidence_ids": [ "cpl-ga4-data" ] }, { "question": "My export has missing dates. What happened?", "answer": "Two documented causes. On the free plan only the first 1,000 rows are exported, which shows up as gaps. The other is an incompatible metric and dimension pair, which returns a 400 error rather than partial data. GA4's per-property API quotas can also bite when several flows pull the same property at once.", "evidence_ids": [ "cpl-ga4-issues" ] }, { "question": "The connector shows as connected but Claude cannot see my data. Why?", "answer": "On individual Claude Free and Pro plans there is a documented defect where the official Coupler.io connector connects but its tools never load, so Claude cannot query the datasets. Coupler.io documents a custom connector URL as a workaround and reports that it works on all plans.", "evidence_ids": [ "cpl-claude-free-pro" ] } ], "related_skills": [], "related_source_ids": [] }, { "id": "google-search-console", "slug": "google-search-console", "name": "Google Search Console", "aliases": [ "GSC", "Search Console", "Webmaster Tools" ], "url": "https://ai-data-connector.com/sources/google-search-console/", "summary": "Google Search Console is Google's free tool for monitoring how a website performs in Google Search. Its data covers queries, clicks, impressions, average position, pages, and URL index status. To analyze it in ChatGPT, Claude, Gemini, or another LLM, use scheduled Coupler.io imports.", "research": { "searched_on": "2026-10-04", "coverage_note": "One route is covered, researched on 4 October 2026: the Coupler.io data platform. No dedicated Search Console entry was reachable in Claude's connector directory at the obvious address; the directory cannot be searched exhaustively, so no claim is made that none exists." }, "evidence": [ { "id": "cpl-claude-destination", "url": "https://docs.coupler.io/destinations/categories/ai/claude", "title": "Claude destination", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-claude-faq", "url": "https://docs.coupler.io/destinations/categories/ai/claude/faq", "title": "Claude destination FAQ", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-claude-issues", "url": "https://docs.coupler.io/destinations/categories/ai/claude/common-issues", "title": "Claude destination common issues", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-gsc-best", "url": "https://docs.coupler.io/sources/category/marketing/google-search-console/best-practices", "title": "Google Search Console best practices", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-gsc-claude-page", "url": "https://www.coupler.io/claude-integrations/google-search-console-to-claude", "title": "Connect Google Search Console to Claude", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-gsc-data", "url": "https://docs.coupler.io/sources/category/marketing/google-search-console/data-overview", "title": "Google Search Console data overview", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-gsc-faq", "url": "https://docs.coupler.io/sources/category/marketing/google-search-console/faq", "title": "Google Search Console source FAQ", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-gsc-issues", "url": "https://docs.coupler.io/sources/category/marketing/google-search-console/common-issues", "title": "Google Search Console source common issues", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-gsc-overview", "url": "https://docs.coupler.io/sources/category/marketing/google-search-console", "title": "Google Search Console source overview", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-mcp", "url": "https://docs.coupler.io/ai/mcp", "title": "What the Coupler.io MCP server can and cannot do", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-refresh", "url": "https://docs.coupler.io/functionality/flow-settings/how-to-set-up-automatic-data-refresh", "title": "How to set up automatic data refresh", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "dir-coupler-io", "url": "https://claude.com/connectors/coupler-io", "title": "Coupler.io in the Claude connectors directory", "publisher": "Anthropic", "checked_on": "2026-10-04", "kind": "directory" } ], "options": [ { "id": "coupler-io", "name": "Coupler.io", "provider": "Coupler.io", "method": "data_platform", "method_label": "Data platform", "maintainer": { "status": "known", "value": "third_party", "evidence_ids": [ "cpl-mcp", "cpl-claude-destination" ], "note": "Coupler.io builds and runs this connector and its MCP server." }, "maintainer_label": "Third-party", "route_status": { "status": "known", "value": "available", "evidence_ids": [ "cpl-claude-destination", "cpl-gsc-claude-page" ] }, "directory_listing": { "status": "known", "value": "https://claude.com/connectors/coupler-io", "evidence_ids": [ "dir-coupler-io" ], "note": "The entry is marked Anthropic verified, credited to Coupler.io, added in November 2025, and records sign-in as required. It lists four tools: get-data, get-schema, list-dataflows, and get-dataflow" }, "links": { "overview": "https://docs.coupler.io/sources/category/marketing/google-search-console" }, "works_with": [ "claude", "chatgpt", "gemini-cli", "gemini-enterprise", "copilot-studio", "perplexity", "cursor", "openclaw", "custom-mcp" ], "access": { "status": "known", "value": "read", "evidence_ids": [ "cpl-mcp", "dir-coupler-io" ], "note": "Queries against imported data are read-only, and nothing in either source describes submitting sitemaps or requesting indexing. Coupler.io's own documentation also describes tools that change your Coupler.io workspace, such as creating a data flow or triggering a run" }, "data_available": { "status": "known", "value": [ "Five report types: Search results performance, Search results performance by appearance, Discover performance, Google News performance, and URLs index performance", "Shared performance metrics: clicks, impressions, ctr, and position", "Dimensions: date, country, device, page, and query, with availability varying by report", "URL inspection columns include verdict, coverage state, robots.txt state, indexing state, last crawl time, page fetch state, Google and user canonical, crawled-as, and AMP, mobile usability, and rich results verdicts", "Filters by dimension support Contains, Does not contain, Equals, Does not equal, Including Regex, and Excluding Regex", "Other options: search results type (Web, News, Image, Video), aggregation (Auto, Page, Property, News showcase panel), and data state (Final or All)" ], "evidence_ids": [ "cpl-gsc-overview", "cpl-gsc-data" ], "note": "Query is available on Search results only, and Device is unavailable on Discover, so the breakdowns you can ask for depend on the report" }, "history": { "status": "known", "value": [ "Search Console retains performance data for approximately 16 months, and nothing older can be queried", "Discover and Google News have a shorter retention period than Search results" ], "evidence_ids": [ "cpl-gsc-data", "cpl-gsc-faq", "cpl-gsc-issues" ], "note": "The documented way to keep longer history is to set up recurring exports and store the results in your destination before they expire" }, "refresh": { "status": "known", "value": [ "A successful manual run is required before a schedule can be set", "Scheduled refresh is described as hourly, daily, or a custom interval", "Coupler.io's data freshness TTL for Search Console is 6 hours, so intra-day refreshes return cached data within that window", "Because GSC data lags by 2 to 3 days, the documented advice is a single daily morning refresh" ], "evidence_ids": [ "cpl-claude-faq", "cpl-refresh", "cpl-gsc-best" ] }, "multi_source": { "status": "known", "value": [ "One data flow can take several sources and combine them with Join or Append before the data reaches Claude", "A worked example joins GSC Search results with GA4 by landing page, noting that GSC uses full URLs while GA4 uses path-only", "GSC clicks and GA4 sessions are counted differently and should be treated as complementary, not equivalent" ], "evidence_ids": [ "cpl-claude-faq", "cpl-gsc-best" ] }, "prerequisites": { "status": "known", "value": [ "A Google Search Console account with verified site ownership or delegated access", "At least Restricted or Full user access to the property", "Owner-level access is not required for performance reports but may be needed for URL inspection data", "A Coupler.io account" ], "evidence_ids": [ "cpl-gsc-overview", "cpl-claude-destination" ] }, "pricing": { "status": "unknown", "value": null, "evidence_ids": [], "note": "no pricing page was checked for this record" }, "data_path": { "status": "known", "value": "Coupler.io reads the Search Console API, stores the result as a dataset, and serves it to Claude over its MCP server. Queries run on Coupler.io's side, so a large dataset does not have to fit into the model's context. Claude sees only datasets from data flows that have Claude as their destination.", "evidence_ids": [ "cpl-mcp" ] }, "limits": { "status": "known", "value": [ "Search Console data is typically delayed by 2 to 3 days; with the Final data state the cutoff is typically 3 to 5 days", "When there are too many unique queries, Google groups low-volume ones under (other) to protect privacy, and there is no workaround", "The URL Inspection API has a daily quota per property; Coupler.io fetches up to 20 URLs concurrently", "CTR and position are averages and are not directly summable across dimensions without recalculation", "Row counts in Claude can differ from the Coupler.io preview when an identifier repeats, so the count should be checked in Claude" ], "evidence_ids": [ "cpl-gsc-data", "cpl-gsc-issues", "cpl-gsc-faq", "cpl-claude-issues" ] }, "capabilities": [], "setup_steps": [ { "text": "In Coupler.io, create a data flow and select Google Search Console as the source, then choose the site.", "evidence_ids": [ "cpl-gsc-overview" ] }, { "text": "Pick one of the five report types, set the report period, and add any dimension filters you need.", "evidence_ids": [ "cpl-gsc-data" ] }, { "text": "Choose {destination} as the destination in the data flow's destination step.", "evidence_ids": [ "cpl-claude-destination" ] }, { "text": "Give the data flow a clear name, because {tool} refers to the dataset by that name.", "evidence_ids": [ "cpl-claude-destination" ] }, { "text": "Run the data flow manually once and wait for it to finish successfully.", "evidence_ids": [ "cpl-claude-destination" ] }, { "text": "Schedule a single daily morning refresh, since the data lags by 2 to 3 days and the freshness TTL is 6 hours.", "evidence_ids": [ "cpl-gsc-best", "cpl-claude-faq" ] } ], "success_check": { "text": "Ask {tool} to list the datasets it can see; the Search Console data flow should appear under the name you gave it. Then compare the row count {tool} reports with the run in Coupler.io, because the preview alone does not confirm what {tool} received.", "evidence_ids": [ "cpl-claude-destination", "cpl-claude-issues" ] } } ], "recommendations": [ { "job": "Tracking search performance beyond Google's 16-month window", "option_id": "coupler-io", "reason": "Search Console deletes performance data after about 16 months, and recurring exports into a stored dataset are the documented way to keep it. Queries then run against that store rather than the API, and the data can be joined with GA4 by landing page.", "evidence_ids": [ "cpl-gsc-faq", "cpl-gsc-best", "cpl-mcp" ] } ], "faq": [ { "question": "How far back can Claude see my search data?", "answer": "About 16 months, which is Search Console's own retention period, and Discover and Google News keep even less. No connector can read past it. If you need longer history, the documented approach is recurring exports stored somewhere before the data expires.", "evidence_ids": [ "cpl-gsc-data", "cpl-gsc-faq", "cpl-gsc-issues" ] }, { "question": "Why is today's data missing?", "answer": "Search Console data is typically delayed by 2 to 3 days, and with the Final data state the cutoff is typically 3 to 5 days. Refreshing more often does not help: Coupler.io's freshness TTL for this source is 6 hours, so intra-day refreshes return cached data, and a single daily morning refresh is the documented advice.", "evidence_ids": [ "cpl-gsc-data", "cpl-gsc-faq", "cpl-gsc-best" ] }, { "question": "What is the \"(other)\" row in my query data?", "answer": "When a property has too many unique queries, Google groups the low-volume ones under (other) to protect privacy and manage data volume. The documentation states there is no workaround, because the underlying data is anonymised by Google before any tool can reach it.", "evidence_ids": [ "cpl-gsc-issues" ] }, { "question": "Can Claude submit a sitemap or request indexing?", "answer": "No. Coupler.io is read-only against Search Console: Claude queries an imported dataset and cannot submit sitemaps or request indexing.", "evidence_ids": [ "cpl-mcp" ] }, { "question": "Why can I not just average the CTR and position columns?", "answer": "Because they are already averages. CTR and position are not directly summable across dimensions without recalculating from clicks and impressions, so asking Claude for an average of the position column will give you a number that does not mean what it looks like.", "evidence_ids": [ "cpl-gsc-data" ] } ], "related_skills": [], "related_source_ids": [] }, { "id": "pipedrive", "slug": "pipedrive", "name": "Pipedrive", "aliases": [ "Pipedrive CRM" ], "url": "https://ai-data-connector.com/sources/pipedrive/", "summary": "Pipedrive is a sales CRM that stores deals, contacts, organizations, and activities. To analyze that data in ChatGPT, Claude, Gemini, or another LLM, use scheduled Coupler.io imports, which load one entity at a time into a queryable dataset.", "research": { "searched_on": "2026-10-04", "coverage_note": "One route is covered, researched on 4 October 2026: the Coupler.io data platform. Its documentation was read in full, covering the data available, prerequisites, access, and limits." }, "evidence": [ { "id": "cpl-claude-destination", "url": "https://docs.coupler.io/destinations/categories/ai/claude", "title": "Claude destination", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-claude-faq", "url": "https://docs.coupler.io/destinations/categories/ai/claude/faq", "title": "Claude destination FAQ", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-claude-free-pro", "url": "https://docs.coupler.io/troubleshooting/claude-connector-tools-dont-load-on-free-and-pro-plans", "title": "Claude connector tools do not load on Free and Pro plans", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-claude-issues", "url": "https://docs.coupler.io/destinations/categories/ai/claude/common-issues", "title": "Claude destination common issues", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-mcp", "url": "https://docs.coupler.io/ai/mcp", "title": "What the Coupler.io MCP server can and cannot do", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-pipedrive-best", "url": "https://docs.coupler.io/sources/category/crm/pipedrive/best-practices", "title": "Pipedrive best practices", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-pipedrive-claude-page", "url": "https://www.coupler.io/claude-integrations/pipedrive-to-claude", "title": "Connect Pipedrive to Claude", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-pipedrive-data", "url": "https://docs.coupler.io/sources/category/crm/pipedrive/data-overview", "title": "Pipedrive data overview", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-pipedrive-issues", "url": "https://docs.coupler.io/sources/category/crm/pipedrive/common-issues", "title": "Pipedrive source common issues", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-pipedrive-overview", "url": "https://docs.coupler.io/sources/category/crm/pipedrive", "title": "Pipedrive source overview", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-refresh", "url": "https://docs.coupler.io/functionality/flow-settings/how-to-set-up-automatic-data-refresh", "title": "How to set up automatic data refresh", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "dir-coupler-io", "url": "https://claude.com/connectors/coupler-io", "title": "Coupler.io in the Claude connectors directory", "publisher": "Anthropic", "checked_on": "2026-10-04", "kind": "directory" } ], "options": [ { "id": "coupler-io", "name": "Coupler.io", "provider": "Coupler.io", "method": "data_platform", "method_label": "Data platform", "maintainer": { "status": "known", "value": "third_party", "evidence_ids": [ "cpl-mcp", "cpl-claude-destination" ], "note": "Coupler.io builds and runs this connector and its MCP server." }, "maintainer_label": "Third-party", "route_status": { "status": "known", "value": "available", "evidence_ids": [ "cpl-claude-destination", "cpl-pipedrive-claude-page" ] }, "directory_listing": { "status": "known", "value": "https://claude.com/connectors/coupler-io", "evidence_ids": [ "dir-coupler-io" ], "note": "The entry is marked Anthropic verified, credited to Coupler.io, added in November 2025, and records sign-in as required. It lists four tools: get-data, get-schema, list-dataflows, and get-dataflow" }, "links": { "overview": "https://docs.coupler.io/sources/category/crm/pipedrive" }, "works_with": [ "claude", "chatgpt", "gemini-cli", "gemini-enterprise", "copilot-studio", "perplexity", "cursor", "openclaw", "custom-mcp" ], "access": { "status": "known", "value": "read", "evidence_ids": [ "cpl-mcp", "dir-coupler-io" ], "note": "Queries against imported data are read-only, and nothing in either source describes writing back to Pipedrive. Coupler.io's own documentation also describes tools that change your Coupler.io workspace, such as creating a data flow or triggering a run, while the directory entry lists only the four read tools" }, "data_available": { "status": "known", "value": [ "Entities: Deals, Persons, Organizations, Activities, Files, Leads, Call logs, Products", "A ninth entity, All deals (BETA), is named in troubleshooting and best practices but is not listed in either entity table", "Per-entity options: a Pipedrive filter ID, a last-modified date range, and column selection", "No Pipedrive reports: in-app analytics and forecast reports are not available through the API, so this is raw CRM data only" ], "evidence_ids": [ "cpl-pipedrive-overview", "cpl-pipedrive-data", "cpl-pipedrive-issues" ], "note": "The overview says nine entities while both entity tables list eight; the count and the tables are not reconciled in the documentation" }, "history": { "status": "unknown", "value": null, "evidence_ids": [], "note": "no retention or backfill window is stated; the only date control documented is a last-modified filter, which can be left empty to export every record" }, "refresh": { "status": "known", "value": [ "A successful manual run is required before a schedule can be set", "Scheduled refresh is described as hourly, daily, or a custom interval", "The platform supports intervals from every 15 minutes to monthly, depending on plan" ], "evidence_ids": [ "cpl-claude-faq", "cpl-refresh" ], "note": "Which intervals each plan allows is not documented" }, "multi_source": { "status": "known", "value": [ "One data flow can take several sources and combine them with Join or Append before the data reaches Claude", "Each data flow exports one entity, so combining Deals with Persons means a flow for each and a join", "Documented join keys: Deals to Persons on person_id.value, and Deals to Organizations on org_id.value" ], "evidence_ids": [ "cpl-claude-faq", "cpl-pipedrive-data", "cpl-pipedrive-best" ] }, "prerequisites": { "status": "known", "value": [ "A Pipedrive account with access to the data you want to export", "The All deals (BETA) entity requires Pipedrive global admin access", "A Coupler.io account" ], "evidence_ids": [ "cpl-pipedrive-overview", "cpl-pipedrive-issues", "cpl-claude-destination" ], "note": "The prerequisites list names no Pipedrive role or plan tier beyond account access; the documentation notes only that some plan levels restrict visibility of other users' activities" }, "pricing": { "status": "unknown", "value": null, "evidence_ids": [], "note": "no pricing page was checked for this record" }, "data_path": { "status": "known", "value": "Coupler.io reads the Pipedrive API, stores the result as a dataset, and serves it to Claude over its MCP server. Queries run on Coupler.io's side, so a large dataset does not have to fit into the model's context. Claude sees only datasets from data flows that have Claude as their destination.", "evidence_ids": [ "cpl-mcp" ] }, "limits": { "status": "known", "value": [ "Pipedrive rate limits are 20 requests per 2 seconds on entry-level plans; Coupler.io paces requests automatically for large accounts, which lengthens run times", "An import that does not finish within 30 minutes times out", "Call logs are fetched at a maximum of 50 records per page, a Pipedrive API constraint", "The Activities entity is fetched with user_id=0, so it returns activities from all users rather than only the connected account", "The Pipedrive API returns active records only; deleted or archived deals, persons, and organizations are excluded", "Date range filtering applies to Deals; for other entities it may have limited or no effect", "Row counts in Claude can differ from the Coupler.io preview when an identifier repeats, so the count should be checked in Claude" ], "evidence_ids": [ "cpl-pipedrive-data", "cpl-pipedrive-issues", "cpl-pipedrive-best", "cpl-claude-issues" ] }, "capabilities": [], "setup_steps": [ { "text": "In Coupler.io, create a data flow and add Pipedrive as the source, then choose the entity you want.", "evidence_ids": [ "cpl-claude-destination", "cpl-pipedrive-data" ] }, { "text": "Narrow the export with a Pipedrive filter ID, a last-modified date range, or a column selection if you do not need every field.", "evidence_ids": [ "cpl-pipedrive-data" ] }, { "text": "Choose {destination} as the destination in the data flow's destination step.", "evidence_ids": [ "cpl-claude-destination" ] }, { "text": "Give the data flow a clear name, because {tool} refers to the dataset by that name.", "evidence_ids": [ "cpl-claude-destination" ] }, { "text": "Run the data flow manually once and wait for it to finish successfully.", "evidence_ids": [ "cpl-claude-destination" ] }, { "text": "Set a refresh schedule once that first run has succeeded.", "evidence_ids": [ "cpl-claude-faq" ] } ], "success_check": { "text": "Ask {tool} to list the datasets it can see; the Pipedrive data flow should appear under the name you gave it. Then compare the row count {tool} reports with the run in Coupler.io, because the preview alone does not confirm what {tool} received.", "evidence_ids": [ "cpl-claude-destination", "cpl-claude-issues" ] } } ], "recommendations": [ { "job": "Analysing sales history at length or alongside other systems", "option_id": "coupler-io", "reason": "Queries run against a stored dataset rather than the CRM, so a long deal history does not have to fit into the conversation, and one data flow can join Deals with Persons or Organizations before Claude sees them.", "evidence_ids": [ "cpl-mcp", "cpl-pipedrive-best" ] } ], "faq": [ { "question": "Can Claude update my CRM, or only read it?", "answer": "Only read it. Coupler.io is read-only against Pipedrive: Claude queries an imported dataset and cannot create or change records.", "evidence_ids": [ "cpl-mcp" ] }, { "question": "Can I get my Pipedrive reports and forecasts into Claude?", "answer": "Not as reports. Pipedrive's in-app analytics and forecast reports are not available through the API, so Coupler.io exports raw CRM data only and you would rebuild the analysis in Claude.", "evidence_ids": [ "cpl-pipedrive-data" ] }, { "question": "Why is my import slow or timing out?", "answer": "Pipedrive rate limits entry-level plans to 20 requests per 2 seconds, and Coupler.io paces its requests automatically for large accounts, which lengthens runs. An import that does not finish within 30 minutes times out. Narrowing the export with a filter ID, a last-modified date range, or fewer columns is the documented way to keep runs short.", "evidence_ids": [ "cpl-pipedrive-issues", "cpl-pipedrive-data" ] }, { "question": "Will an import include deals I have deleted or archived?", "answer": "No. The Pipedrive API returns active records only, so deleted and archived deals, persons, and organizations are excluded from a Coupler.io import.", "evidence_ids": [ "cpl-pipedrive-best" ] }, { "question": "Does the activity data cover my whole team?", "answer": "Yes, and that may surprise you. Coupler.io fetches the Activities entity with user_id=0, which returns activities from all Pipedrive users rather than only the connected account. Separately, some Pipedrive plan levels restrict visibility of other users' activities.", "evidence_ids": [ "cpl-pipedrive-data", "cpl-pipedrive-issues" ] }, { "question": "The connector shows as connected but Claude cannot see my data. Why?", "answer": "On individual Claude Free and Pro plans there is a documented defect where the official Coupler.io connector connects but its tools never load, so Claude cannot query the datasets. Coupler.io documents a custom connector URL as a workaround and reports that it works on all plans.", "evidence_ids": [ "cpl-claude-free-pro" ] } ], "related_skills": [], "related_source_ids": [] }, { "id": "quickbooks-online", "slug": "quickbooks-online", "name": "QuickBooks Online", "aliases": [ "QBO", "QuickBooks", "Intuit QuickBooks" ], "url": "https://ai-data-connector.com/sources/quickbooks-online/", "summary": "QuickBooks Online is Intuit's cloud accounting software. It stores invoices, customers, vendors, transactions, and financial reports such as profit and loss and A/R aging. To analyze that data in ChatGPT, Claude, Gemini, or another LLM, use scheduled Coupler.io imports.", "research": { "searched_on": "2026-10-04", "coverage_note": "One route is covered, researched on 4 October 2026: the Coupler.io data platform. Its documentation was read in full, covering the data available, prerequisites, access, and limits." }, "evidence": [ { "id": "cpl-claude-destination", "url": "https://docs.coupler.io/destinations/categories/ai/claude", "title": "Claude destination", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-claude-faq", "url": "https://docs.coupler.io/destinations/categories/ai/claude/faq", "title": "Claude destination FAQ", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-claude-free-pro", "url": "https://docs.coupler.io/troubleshooting/claude-connector-tools-dont-load-on-free-and-pro-plans", "title": "Claude connector tools do not load on Free and Pro plans", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-claude-issues", "url": "https://docs.coupler.io/destinations/categories/ai/claude/common-issues", "title": "Claude destination common issues", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-mcp", "url": "https://docs.coupler.io/ai/mcp", "title": "What the Coupler.io MCP server can and cannot do", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-qbo-claude-page", "url": "https://www.coupler.io/claude-integrations/quickbooks-to-claude", "title": "Connect QuickBooks to Claude", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-qbo-data", "url": "https://docs.coupler.io/sources/category/finance-and-accounting/quickbooks/data-overview", "title": "QuickBooks data overview", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-qbo-faq", "url": "https://docs.coupler.io/sources/category/finance-and-accounting/quickbooks/faq", "title": "QuickBooks source FAQ", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-qbo-issues", "url": "https://docs.coupler.io/sources/category/finance-and-accounting/quickbooks/common-issues", "title": "QuickBooks source common issues", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-qbo-overview", "url": "https://docs.coupler.io/sources/category/finance-and-accounting/quickbooks", "title": "QuickBooks source overview", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "cpl-refresh", "url": "https://docs.coupler.io/functionality/flow-settings/how-to-set-up-automatic-data-refresh", "title": "How to set up automatic data refresh", "publisher": "Coupler.io", "checked_on": "2026-10-04", "kind": "vendor_docs" }, { "id": "dir-coupler-io", "url": "https://claude.com/connectors/coupler-io", "title": "Coupler.io in the Claude connectors directory", "publisher": "Anthropic", "checked_on": "2026-10-04", "kind": "directory" } ], "options": [ { "id": "coupler-io", "name": "Coupler.io", "provider": "Coupler.io", "method": "data_platform", "method_label": "Data platform", "maintainer": { "status": "known", "value": "third_party", "evidence_ids": [ "cpl-mcp", "cpl-claude-destination" ], "note": "Coupler.io builds and runs this connector and its MCP server." }, "maintainer_label": "Third-party", "route_status": { "status": "known", "value": "available", "evidence_ids": [ "cpl-claude-destination", "cpl-qbo-claude-page" ] }, "directory_listing": { "status": "known", "value": "https://claude.com/connectors/coupler-io", "evidence_ids": [ "dir-coupler-io" ], "note": "The entry is marked Anthropic verified, credited to Coupler.io, added in November 2025, and records sign-in as required. It lists four tools: get-data, get-schema, list-dataflows, and get-dataflow" }, "links": { "overview": "https://docs.coupler.io/sources/category/finance-and-accounting/quickbooks" }, "works_with": [ "claude", "chatgpt", "gemini-cli", "gemini-enterprise", "copilot-studio", "perplexity", "cursor", "openclaw", "custom-mcp" ], "access": { "status": "known", "value": "read", "evidence_ids": [ "cpl-mcp", "dir-coupler-io" ], "note": "Queries against imported data are read-only, and nothing in either source describes writing back to QuickBooks Online. Coupler.io's own documentation also describes tools that change your Coupler.io workspace, such as creating a data flow or triggering a run, while the directory entry lists only the four read tools; the two sources disagree on how much of that workspace control the connector currently exposes" }, "data_available": { "status": "known", "value": [ "Entities: Invoice, Customer, Bill, Payment, Vendor, Employee, Estimate, Purchase, PurchaseOrder, SalesReceipt, CreditMemo, Deposit, Transfer, JournalEntry, Account, Item, Budget, TimeActivity", "Further entities: Attachable, BillPayment, Class, CompanyInfo, Department, ExchangeRate, JournalCode, PaymentMethod, Preferences, RefundReceipt, TaxAgency, TaxCode, TaxRate, Term, VendorCredit", "Summary reports: Profit and Loss Summary, Balance Sheet, Cash Flow, Trial Balance, Expenses by Vendor, Sales by Customer, Sales by Product, Sales by Class Summary, Sales by Department, Tax Summary, Inventory Valuation Summary", "Detail reports: Profit and Loss Detail, General Ledger Detail, Transaction List, Transaction List by Customer, Transaction List by Vendor, Transaction List with Splits, Journal Report", "Aging and balance reports: AP Aging Detail and Summary, AR Aging Detail and Summary, Customer Balance and Detail, Vendor Balance and Detail, Customer Income, Account List Detail" ], "evidence_ids": [ "cpl-qbo-data" ], "note": "Raw entities and pre-built reports are two separate sources; one entity or report per data flow" }, "history": { "status": "unknown", "value": null, "evidence_ids": [], "note": "no retention or backfill window is stated; the documented constraint is the size of a single report rather than how far back it can reach" }, "refresh": { "status": "known", "value": [ "A successful manual run is required before a schedule can be set", "Scheduled refresh is described as hourly, daily, or a custom interval", "The platform supports intervals from every 15 minutes to monthly, depending on plan" ], "evidence_ids": [ "cpl-claude-faq", "cpl-refresh" ], "note": "Which intervals each plan allows is not documented" }, "multi_source": { "status": "known", "value": [ "One data flow can take several sources and combine them with Join or Append before the data reaches Claude", "Within Claude, a SQL dataset can be built from datasets in the same data flow, because one query reads one dataset" ], "evidence_ids": [ "cpl-claude-faq", "cpl-mcp" ] }, "prerequisites": { "status": "known", "value": [ "A QuickBooks Online account on Simple Start, Essentials, Plus, or Advanced", "Admin or Accountant access to the company file", "Classes, Locations, and Budgets require the Plus or Advanced tier", "A Coupler.io account" ], "evidence_ids": [ "cpl-qbo-overview", "cpl-qbo-faq", "cpl-claude-destination" ] }, "pricing": { "status": "unknown", "value": null, "evidence_ids": [], "note": "no pricing page was checked for this record" }, "data_path": { "status": "known", "value": "Coupler.io reads the QuickBooks Online API, stores the result as a dataset, and serves it to Claude over its MCP server. Queries run on Coupler.io's side, so a large dataset does not have to fit into the model's context. Claude sees only datasets from data flows that have Claude as their destination.", "evidence_ids": [ "cpl-mcp" ] }, "limits": { "status": "known", "value": [ "The QuickBooks Reports API truncates any report above 400,000 cells (rows times columns); this is Intuit's limit and cannot be raised", "QuickBooks Desktop, including Desktop Enterprise and Desktop Pro, is not supported", "The QuickBooks query language supports AND but not OR, and reports can be grouped by only one dimension at a time", "Custom reports built in the QuickBooks interface are not available through the API", "Entity exports return only active records by default", "Row counts in Claude can differ from the Coupler.io preview when an identifier repeats, so the count should be checked in Claude" ], "evidence_ids": [ "cpl-qbo-data", "cpl-qbo-faq", "cpl-qbo-issues", "cpl-claude-issues" ] }, "capabilities": [ { "text": "Ask questions about imported QuickBooks entities and pre-built reports, such as invoices, bills, customers, profit and loss, balance sheet, and A/R and A/P aging.", "evidence_ids": [ "cpl-qbo-data", "cpl-mcp" ] }, { "text": "Keep the data current with a refresh schedule once a first manual run has succeeded.", "evidence_ids": [ "cpl-claude-faq", "cpl-refresh" ] }, { "text": "Combine QuickBooks with other sources in one data flow, using Join or Append, before Claude sees the result.", "evidence_ids": [ "cpl-claude-faq", "cpl-mcp" ] }, { "text": "Run SQL against a dataset on Coupler.io's side, so a large dataset does not have to fit into Claude's context.", "evidence_ids": [ "cpl-mcp" ] } ], "setup_steps": [ { "text": "In Coupler.io, create a data flow and add QuickBooks as the source, choosing either an entity or a pre-built report.", "evidence_ids": [ "cpl-claude-destination", "cpl-qbo-data" ] }, { "text": "Choose {destination} as the destination in the data flow's destination step.", "evidence_ids": [ "cpl-claude-destination" ] }, { "text": "Give the data flow a clear name, because {tool} refers to the dataset by that name.", "evidence_ids": [ "cpl-claude-destination" ] }, { "text": "Run the data flow manually once and wait for it to finish successfully.", "evidence_ids": [ "cpl-claude-destination" ] }, { "text": "Set a refresh schedule once that first run has succeeded.", "evidence_ids": [ "cpl-claude-faq" ] } ], "success_check": { "text": "Ask {tool} to list the datasets it can see; the QuickBooks data flow should appear under the name you gave it. Then compare the row count {tool} reports with the run in Coupler.io, because the preview alone does not confirm what {tool} received.", "evidence_ids": [ "cpl-claude-destination", "cpl-claude-issues" ] } } ], "recommendations": [ { "job": "Analysing QuickBooks figures alongside other systems", "option_id": "coupler-io", "reason": "One data flow can join or append several sources before Claude sees them, and queries run against the stored dataset rather than the QuickBooks API, so a large report does not have to fit into the conversation.", "evidence_ids": [ "cpl-claude-faq", "cpl-mcp" ] } ], "faq": [ { "question": "Does this work with QuickBooks Desktop?", "answer": "No. Coupler.io connects to the QuickBooks Online API only, and the documentation states that Desktop, including Desktop Enterprise and Desktop Pro, is not supported. Moving to QuickBooks Online would be a prerequisite.", "evidence_ids": [ "cpl-qbo-faq", "cpl-qbo-overview" ] }, { "question": "Can Claude create or send an invoice, or only read my books?", "answer": "Only read them. Coupler.io is read-only against QuickBooks: Claude queries an imported dataset, and nothing in its documentation describes writing back to your books.", "evidence_ids": [ "cpl-mcp" ] }, { "question": "Which Claude apps can read the data once it is connected?", "answer": "Coupler.io's documentation names Claude web, Claude desktop, mobile, Cowork, and Claude Code, and says a single web connector covers them; Claude Code is connected with the /mcp command instead. The connector only needs to be installed once.", "evidence_ids": [ "cpl-claude-destination", "cpl-claude-faq" ] }, { "question": "Can Claude change my accounting data?", "answer": "Queries against the imported data are documented as read-only, so the model cannot edit, delete, or overwrite imported rows. The same connection does expose tools that can change your Coupler.io workspace, such as creating a data flow or triggering a run, and the documentation says the model is instructed to confirm before making a change you did not ask for.", "evidence_ids": [ "cpl-mcp" ] }, { "question": "Why can my reports get cut off?", "answer": "The QuickBooks Reports API truncates a report once it exceeds 400,000 cells, counting rows times columns. This is Intuit's limit, and the documentation says there is no way to raise it, so a long date range on a wide report has to be split up.", "evidence_ids": [ "cpl-qbo-data", "cpl-qbo-issues" ] }, { "question": "The connector shows as connected but Claude cannot see my data. Why?", "answer": "On individual Claude Free and Pro plans there is a documented defect where the official connector connects but its tools never load, so Claude cannot query the datasets. Coupler.io documents a custom connector URL as a workaround and reports that it works on all plans.", "evidence_ids": [ "cpl-claude-free-pro" ] } ], "related_skills": [], "related_source_ids": [] } ] }