OtterlyAI MCP server for your AI

No. The connect form asks for one field, API key, and you paste the value copied from OtterlyAI. There is no OAuth app, client secret, or redirect URL to register anywhere. Once the key is saved, PopMCP hosts the server and your AI client calls it at a URL.

30 operations across 7 API modules, split 22 read and 8 write. All 30 load by default, so nothing sits behind an extra search step. Brand reports is the largest module with 11 operations, followed by audits with 9.

Eight operations write: create_prompts, delete_prompt, update_prompt_tags, create_tag, delete_tag, create_content_check, create_crawlability_check and create_query_fan_out. Everything else, including brand reports, stats, citations, AI answers and recommendations, only reads. Those writes run as soon as the AI calls them; for pure reporting, keep teammates on read-only.

Listing and creating prompts and tags happens inside one workspace. list_workspaces returns the workspaces on the account, and list_workspace_prompts, list_workspace_tags, create_prompts and create_tag each act inside one of them. Name the workspace in your request, or have the agent list them and confirm before it writes.

list_engines returns the AI engines and the countries they support, so the agent can check coverage rather than assume it. get_account_info returns account information, and plan usage is part of what the API covers. Worth reading that before you point an agent at a long backfill of brand reports.

Each is a create-then-read pair. create_crawlability_check, create_content_check and create_query_fan_out start the job, while get_crawlability_check, get_content_check and get_query_fan_out read one back by id, with list_crawlability_checks, list_content_checks and list_query_fan_outs for history. Have the agent reuse the id from the create call when it reads the result, rather than starting a second check for the same URL.

No. PopMCP hosts it and gives you a Streamable HTTP MCP URL. There is nothing to deploy, no callback URL to register, and no infrastructure to keep running or patch.

Any client that supports a remote Streamable HTTP MCP server: Claude Desktop, Claude Web, Claude Code, ChatGPT custom connectors, Codex and the Codex CLI, OpenCode and Z.ai's ZCode, as well as editors such as Cursor, Windsurf and Zed. Your client signs in to PopMCP with OAuth; the provider credential never leaves the server.

Connect OtterlyAI in PopMCP first, then run claude mcp add --transport http popmcp https://app.popmcp.com/mcp in a terminal. Inside Claude Code, run /mcp and sign in to PopMCP in the browser window it opens, ticking OtterlyAI on the consent screen. The OtterlyAI tools are then available in that session, and no OtterlyAI credential is stored on your machine.

Yes. The OtterlyAI catalog is small enough that all 30 tool definitions load into every conversation, so your AI never has to search for one before it can call it.

A tool is something your AI calls, and there are 30 here. A trigger is something PopMCP watches for you: 7 OtterlyAI events that can start work on the PopMCP side. Triggers never enter your AI client's context, and none of them ask you to paste a callback URL into OtterlyAI.