Integrations

MCP server

Let coding agents install the SDK for you and query your analytics over MCP.

AgentStats runs an MCP server at https://agentstats.io/mcp (Streamable HTTP). Connect it to Claude Code, Cursor or any MCP client so your coding agent can install the SDK for you, or answer questions like "which tools failed most this week?". Every tool only reads, except update_recommendation, which records that a recommendation was done, and run_evals, which starts eval runs.

Connect

claude.ai, ChatGPT and other connectors: add https://agentstats.io/mcp as a custom connector. When a tool needs your data, the connector asks you to sign in to AgentStats and allow read access for your team (OAuth, as the MCP spec describes). You can revoke it any time by signing out of the connector.

Claude Code, Cursor and clients that take headers: create a team API key under Account settings → API keys (al_sk_…), then add the server to your client:

claude mcp add --transport http agentstats https://agentstats.io/mcp --header "Authorization: Bearer al_sk_…"

Signing in, or a key, is optional. Without either you only get the setup tools.

Tools

ToolNeeds a keyWhat it does
get_setup_instructionsNoInstall steps for a given framework. With a key, they include your real public keys
score_webmcp_toolsNoScore tool definitions 0–100 for agent-readiness, with concrete issues
list_sitesYesYour team's websites, with public keys and tool counts
get_overviewYesCalls, sessions, success rate and latency with trends, top tools and recent errors
get_tool_detailYesOne tool in depth: latency, error groups, recent calls, schema history
get_errorsYesFailed calls grouped by tool and error message
get_schemaYesThe live tool surface, its readiness score and recent schema changes
run_funnelYesDrop-off across a sequence of 2–8 tools
list_sessionsYesRecent agent sessions and the tools each one called
get_visitorsYesThe people on the website: visitors, sources, UTM tags, pages, countries, devices and goals, with filters
get_trafficYesHuman visitors vs. agent sessions by day, plus AI-assistant referrals
get_seoYesGoogle and Bing search: clicks, top queries and pages, indexing suggestions and page speed
get_ai_botsYesAI crawler and agent visits by product and category
get_ai_visibilityYesWhat ChatGPT, Claude, Gemini, Perplexity and Google AI Mode say about your product, from the latest AI visibility run
get_recommendationsYesEverything worth doing in the project, ranked, with each one's status (open, planned)
update_recommendationYesMarks a recommendation planned, done or not for us (or back to open), after you've acted on it
get_agent_browsersYesAgents browsing your pages in a real browser (Claude in Chrome, automation)
get_docs_trafficYesHow agents read your docs: Markdown served, most-read pages, pages that don't exist
get_mcp_sign_inYesOAuth sign-ins on your MCP servers, by client app and scope
get_feedbackYesWhat agents and users sent through an MCP server's send_feedback tool
list_evalsYesAn MCP server's evals with each agent's latest verdict
run_evalsYesStarts an MCP server's evals against the live server
get_eval_runYesOne eval run: verdict, score, problems with fixes, every tool call

llms.txt

For agents that don't speak MCP, /llms.txt has the same setup instructions as plain Markdown. Paste the link into any chat and ask it to install AgentStats.