Google as tools for your agent. Unofficial: not affiliated with or endorsed by google.com. Unbrowse compiled these from google.com's own requests: each call replays the site's first-party API server-side and returns a verified result. No scraping of rendered pages, no browser on your side, no google.com API key.
Connect the google.com MCP
This server is a tool scope holding only google.com: its compiled tools plus the Unbrowse tools that run them, so the agent's tool list stays small and every call stays on google.com.
claude mcp add --transport http google https://unbrowse.ai/mcp/google.com
codex mcp add google --url https://unbrowse.ai/mcp/google.com && codex mcp login google
{"mcpServers":{"google":{"url":"https://unbrowse.ai/mcp/google.com"}}}
The first connect signs in to Unbrowse with OAuth (free tier, only verified results count). For scripts, send Authorization: Bearer $UNBROWSE_API_KEY (key from https://unbrowse.ai/app/keys; keep it in a secret manager, never in config you commit).
If none of the tools below are in this session's tool list, the server is not connected or not signed in: say so and stop. Do not fetch google.com some other way and present it as this skill's result.
Tools
google_com__get_travel_flights — Google Flights SIN-BKK fare compare
Google Flights SIN-BKK fare compare. No inputs. Returns the page's title, readable text and links. Read-only on google.com.
q(string) — optional, e.g. "Flights from Singapore to Bangkok on 2026-11-14"
{"q":"Flights from Singapore to Bangkok on 2026-11-14"}
google_com__get_complete_s — Search Google results
Search Google results. Inputs: q; gs_pcrt. Returns text. Read-only on google.com.
q(string, required)gs_pcrt(integer, required) — gs pcrtsei(string) — optional, e.g. "SBi5aoHBMq_gseMPzoHt8QY"cp(integer) — optional, e.g. "0"client(string) — optional, e.g. "gws-wiz-serp"xssi(string) — optional, e.g. "t"authuser(integer) — optional, e.g. "0"pq(string) — optional, e.g. "guatemala capital"psi(string) — optional, e.g. "SRi5at-eAc2fseMP6ZCTsQU.1790515274369"dpr(integer) — optional, e.g. "1"nolsbt(integer) — optional, e.g. "1"- 1 more of the site's own parameters (locale, paging and the like), sent as recorded
{"q":"<q>","gs_pcrt":"<gs_pcrt>"}
google_com__get_preview_place — Open a place page on Google Maps
Open a place page on Google Maps. No inputs. Returns text. Read-only on google.com.
authuser(integer) — optional, e.g. "0"q(string) — optional, e.g. "Jamaica Defence Force Co-Operative Credit Union Limited"
{"authuser":0,"q":"Jamaica Defence Force Co-Operative Credit Union Limited"}
google_com__render_page — Get directions on Google Maps
Get directions on Google Maps. Inputs: origin. Returns the page's title, readable text and links. Read-only on google.com.
origin(string, required) — origin — what to search for on google.com
{"origin":"<origin>"}
google_com__get_s — Search places on Google Maps
Search places on Google Maps. Inputs: query. Returns text. Read-only on google.com.
query(string, required) — query (typed during “fill Search Google Maps”)tbm(string) — optional, e.g. "map"gs_ri(string) — optional, e.g. "maps"suggest(string) — optional, e.g. "p"authuser(integer) — optional, e.g. "0"psi(string) — optional, e.g. "9v-5aouUM8PF1e8PuqvU8Qk.1790574592065.1"ech(integer) — optional, e.g. "1"
{"query":"<query>"}
news_google_com__get_rss_search — Google News search (RSS)
Google News search (RSS). Inputs: query. Returns text. Read-only on news.google.com.
query(string, required) — query (typed during “search”)ceid(string) — optional, e.g. "US:en"
{"query":"<query>"}
business_google_com__get_resources_search — Search thinkwithgoogle.com
Read 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.
query(string, required) — query (typed during “fill Search”), e.g. "article", "description"
{"query":"article"}
google_com__read_page — Read a page on google.com
Read 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.
path(string, required) — A page on google.com: a path like /about, or a full URL on google.com
{"path":"/"}
Always there too: unbrowse.run (a task on google.com in plain words), unbrowse.inspect, unbrowse.resume, unbrowse.scrape and unbrowse.map (any page on google.com), and the recorded cloud browser unbrowse.browse.open → unbrowse.browse.act → unbrowse.browse.finish. Hosts that reject dotted names show them with _ (unbrowse_run).
How to call
- Pick the tool whose title matches the job and pass its inputs. Leave the site's recorded parameters out unless the user asked for something they control.
- Read
status.succeededcarries the verifiedresult.input_requiredis not a failure: answer the open fields withunbrowse.resume {runId, answers}on the same run. - No tool fits:
unbrowse.run {task}. It fails withno_capabilitywhen nothing is compiled for that yet: do it once withunbrowse.browse.openon google.com →unbrowse.browse.act→unbrowse.browse.finish, which compiles a new tool this server lists next time. Anything off google.com is refused here (not_in_scope): use the general Unbrowse server for that. - Signing in: never ask the user for a password.
unbrowse.browse.actwithaction: "autofill"fills google.com's login from their Unbrowse password manager. - Report only what a
succeededresult says.outcome_unknownmeans a change may have happened: check before retrying. Posting, buying, sending or deleting needs the user's go-ahead.
Without MCP
Every tool is also a REST call:
curl -s https://unbrowse.ai/api/v1/sites/google.com/call/google_com__get_travel_flights \
-H "authorization: Bearer $UNBROWSE_API_KEY" -H "content-type: application/json" \
-d '{"q":"Flights from Singapore to Bangkok on 2026-11-14"}'
OpenAPI: https://unbrowse.ai/api/v1/sites/google.com/openapi.json. Every other site: the general Unbrowse skill (https://github.com/unbrowse-ai/unbrowse) and https://unbrowse.ai/mcp.