How much does Google Super cost?

Google Super uses a free pricing model.

What are the main advantages of Google Super?

Massive tool library with over 200 specific operations across the entire Google suite.. Specifically optimized for the Model Context Protocol (MCP), ensuring high compatibility with modern AI agent frameworks.. Enables autonomous, multi-step agentic workflows that cross-reference data between different Google services (e.g., Sheets to Gmail).. Streamlines authentication by providing a unified OAuth gateway via platforms like Composio or Smithery.. High level of agent autonomy, allowing for 'compound' tasks like searching for a file, editing it, and sending it as an attachment..

What are the drawbacks of Google Super?

Large security surface area; researchers highlight that 200+ tools are 'unauditable' for most users and increase the risk of accidental data destruction.. Initial configuration can be technical and error-prone, particularly regarding OAuth scopes and environment variables.. Agent performance can degrade when overwhelmed by too many tool options (context window competition).. Opaque error handling in some instances, making it difficult to debug why a specific agentic action failed..

What are the best alternatives to Google Super?

Alternatives to Google Super in this category include Linkup, Google Sheets, Slack, Gmail, AgentMail.

Google Super

Unclaimed verified 28 oct 2026

[ score · 30 ]

Pricing: Free Last verified: 2026-10-28

Visit Website

Updated Oct 29, 2026

TL;DR

Google Super is an all-in-one Model Context Protocol (MCP) server and toolkit that consolidates over 200 Google API services—including Gmail, Drive, Calendar, Sheets, and Ads—into a unified interface for AI agents. It is primarily designed for developers and AI power users who want to give LLMs like Claude or GPT-4 direct, autonomous control over their digital workspace. Its key differentiator is the massive breadth of integrated 'tools' available via a single connection, eliminating the need for individual API setups.

What Users Actually Pay

No user-reported pricing yet.

Our Take

Google Super occupies a specialized but critical niche in the burgeoning AI agent ecosystem. By wrapping the fragmented Google Cloud API landscape into a standardized MCP surface, it allows agents to perform complex, multi-app workflows—such as analyzing a spreadsheet, drafting a summary in Docs, and emailing it—without manual intervention. It effectively bridges the gap between static LLMs and the 'agentic' execution of tasks within the Google ecosystem. The product's greatest strength is its comprehensiveness; it doesn't just stop at Gmail and Drive but includes deeper integrations like Google Ads, Analytics, and Photos. This makes it a powerful choice for marketing and data analysis agents. However, this 'super' approach is also its primary liability. Security researchers have pointed out that exposing 200+ tools, many with destructive capabilities (like deleting files or emails), creates a significant 'unauditable' risk if the agent's instructions are not strictly governed. Best suited for developers building autonomous AI workers or users of advanced AI desktops (like Claude Desktop or Cursor) who require deep Google integration. It is less a consumer 'app' and more a backend facilitator for the next generation of AI productivity. While powerful, users must be prepared for the complexities of OAuth configuration and the potential for agent errors given the high number of available tools.

Alternatives

Ranked by Revuo score — paid tiers never affect order.

Linkup

Google Sheets

Slack

Gmail

AgentMail

GitHub

Pros

+ Massive tool library with over 200 specific operations across the entire Google suite.
+ Specifically optimized for the Model Context Protocol (MCP), ensuring high compatibility with modern AI agent frameworks.
+ Enables autonomous, multi-step agentic workflows that cross-reference data between different Google services (e.g., Sheets to Gmail).
+ Streamlines authentication by providing a unified OAuth gateway via platforms like Composio or Smithery.
+ High level of agent autonomy, allowing for 'compound' tasks like searching for a file, editing it, and sending it as an attachment.

Cons

- Large security surface area; researchers highlight that 200+ tools are 'unauditable' for most users and increase the risk of accidental data destruction.
- Initial configuration can be technical and error-prone, particularly regarding OAuth scopes and environment variables.
- Agent performance can degrade when overwhelmed by too many tool options (context window competition).
- Opaque error handling in some instances, making it difficult to debug why a specific agentic action failed.

Agent Readiness

65/100

Google Super is exceptionally ready for AI agents, as it is natively built for the Model Context Protocol (MCP). It provides a massive surface area of 200+ tools, enabling high autonomy for agents to manage digital lives. Integration with orchestration platforms like Composio extends its reach to Zapier, n8n, and various LLM frameworks, though developers must manage the security risks associated with such broad permissions.

API Surface100

Public APIRESTMCPFree TieropenApi

Protocol Support40

MCP (200 tools)

SDK Availability0

Integration Ecosystem100

ZapierMaken8nWebhooksClaude DesktopCursorOpenAI Assistant APILangChainCrewAI

Developer Experience65

Docs: goodVersioningChangelog

Last checked May 9, 2026

MCP Integrations

2 servers27 tools954 total uses

Google Scholargoogle/scholar

smitheryRemoteHigh match

Search academic papers, citations, and case law from Google Scholar. No API key required. Data via SerpAPI; Smithery is not affiliated with Google.

954 uses

1 tool

search_papersSearch Google Scholar for academic papers, books, theses, and citations. Returns organic results with title, authors, snippet, citation count, and PDF links.

Unbrowse for Google (unofficial)unbrowse/google

smitheryRemoteHigh match

google.com as MCP tools: Google News search (RSS). Compiled by Unbrowse from the site's own requests; verified results, no browser. Not affiliated with google.com.

26 tools

google_com__get_travel_flightsGoogle Flights SIN-BKK fare compare. Compiled from google.com's own requests and shared by an Unbrowse user. No inputs. Returns the page's title, readable text and links. Read-only on google.com.
google_com__get_complete_sSearch Google results. Compiled from google.com's own requests and indexed by Unbrowse. Inputs: q; gs_pcrt. Returns text. Read-only on google.com.
google_com__get_preview_placeOpen a place page on Google Maps. Compiled from google.com's own requests and indexed by Unbrowse. No inputs. Returns text. Read-only on google.com.
google_com__render_pageGet directions on Google Maps. Compiled from google.com's own requests and indexed by Unbrowse. Inputs: origin. Returns the page's title, readable text and links. Read-only on google.com.
google_com__get_sSearch places on Google Maps. Compiled from google.com's own requests and indexed by Unbrowse. Inputs: query. Returns text. Read-only on google.com.
news_google_com__get_rss_searchGoogle News search (RSS). Compiled from news.google.com's own requests and shared by an Unbrowse user. Inputs: query. Returns text. Read-only on news.google.com.
business_google_com__get_resources_searchRead resources search on business.google.com on Think with Google - Marketing Research, Insights, and Trends (business.google.com). Use for requests like “search business.google.com for article”; “search business.google.com for description”; “search business.google.com for description — Search thinkwithgoogle.com for description”. Learned from 2 browser traces; chains get_resources_search. Inputs: query e.g. "article". Returns resources, nextPageToken. Read-only on business.google.com.
google_com__read_pageRead any page on google.com — a path such as /news/2026/some-story, or a full google.com URL — and get its title, readable text and links. One first-party HTTP request, no browser. Follow a returned link by passing its href back as the path. Inputs: path e.g. "/". Returns the page's title, readable text and links. Read-only on google.com.
unbrowse.runStart a durable Unbrowse run by capability id or natural-language task. For a learned capability, pass `input` keyed by the inputs unbrowse.discover lists (e.g. {origin, destination, date, passenger}); anything missing is asked for on the same run. Fails with no_capability when nothing fits. result.next.tool is unbrowse.index (POST /api/v1/index). unbrowse.browse.open is only for a session whose tool list includes it. It is not a REST route, and the CLI does not open a browser.
unbrowse.inspectInspect a run: status, requirements, verified result, known effects, routing explanation.
unbrowse.resumeAnswer the run's open requirements and continue the same run. Simplest: {runId, answers: {<affectedAction>: value}} — e.g. {"origin": "CDG"} or, for a choice, {"selected_offer": <an option's value>}. Answer several at once.
unbrowse.cancelCancel a run. Returns a receipt of dispatched, uncertain and unstarted effects.
unbrowse.browse.openOpen a cloud browser (patchright) on a URL when no capability fits yet. Recording is on from the first navigation, so doing the task here fulfils it now and teaches Unbrowse the site. Returns a snapshot of interactive elements with @refs.
unbrowse.browse.snapshotCurrent page text and interactive elements with @refs. Refs change after every action; always act on the latest snapshot.
unbrowse.browse.actclick | fill | select | check | press | wait on a @ref, or autofill. For fill, pass `name` as the business field (origin, date, email…) so the learned API gets readable inputs. For logins never ask the user for a password: `autofill` fills the page's login form (username or email, password, 2FA code) from the user's Unbrowse password manager, or fill one @ref with `vault: "username" | "email" | "password" | "totp"`. The values go straight into the page and never pass through you. If no login is saved for the site, the error carries a link: give it to the user, wait with unbrowse.credentials.status, then repeat the action.
unbrowse.credentials.listThe user's saved logins (Unbrowse password manager): site, label, masked username/email hints and which fields are saved. Never values — logins are filled into pages with unbrowse.browse.act (autofill, or vault on a field) and used by learned capabilities automatically.
unbrowse.credentials.requestAsk the user to save a login you need, without seeing it: returns a one-time link to give them. They open it, sign in to Unbrowse and save the username/email, password and optional 2FA seed. Then call unbrowse.credentials.status.
unbrowse.credentials.statusWhether the user has saved the login for a request (pending, fulfilled, declined, expired). Waits up to `waitSeconds` (default 20, max 50) for it to change.
unbrowse.browse.finishEnd the session. Returns the page the task ended on and compiles every recorded session for this site into a learned.* capability you can call with unbrowse.run next time — no browser.
unbrowse.browse.closeEnd a browser session. It is still indexed (passive learning) unless discard is true.
unbrowse.scrapeRead one page as clean markdown (main content only by default — no navigation, cookie banners or footers), with its links and metadata (title, description, canonical, language, status). Plain HTTP when the server's HTML has the content; otherwise Unbrowse's cloud browser renders it, signed in when a session is kept for the site. Use it to read a page you have a URL for; use unbrowse.run for a task on a site. One verified call.
unbrowse.mapList a site's URLs: the sitemaps its robots.txt names (or /sitemap.xml, following one level of sitemap indexes), then the links on the given page. Same-site only. Filter with search. Use it to find the right page before scraping it. One verified call.
unbrowse.sitesWhat Unbrowse knows about each website this workspace has touched, before you act: access (public, or only behind a sign-in), the kept signed-in session (active, expired, logged_out or none — and when), the last sign-in, whether a login is saved, the tools already learned there and public tools, and any bot check that blocked it. Check it first: an active session means no sign-in is needed; a learned tool means no browsing is needed. Pass a site for one, or nothing for all.
unbrowse.indexIndex a website on purpose: Unbrowse's own agent opens the site in a cloud browser, performs its core read-only capabilities (search, listings, item pages, filters…), compiles each into a tool and proves it with a browserless replay. Runs in the background (minutes); returns a jobId for unbrowse.index.status. The tools land in your tool list (and, when shared, in the public registry). Use it before a user needs a site, so their first call is instant. Never signs in or changes anything on the site.
unbrowse.index.statusAn index job's progress: queued, running, done or failed, the steps taken and every tool indexed (inputs, whether it runs without a browser, verified, public id). Waits up to `waitSeconds` (default 20, max 50) for it to finish. Without a jobId, lists your recent jobs.
unbrowse.discoverSearch the caller's private space (connected public and passworded sites as flat primitives) and the public registry (second priority, connect-on-the-fly). Each capability carries `hints.health` (healthy, degraded, untested, failing, ...) from its recent runs. Public tools that are failing are left out unless the query names their site; `withheld.count` says how many, and for those sites unbrowse.browse.open learns a fresh route.