MCP server for advanced web search using Tavily
Serpbridge is installed from its publisher's own source and answers where it runs, so this marketplace is not in the path of a single call. There is no address here to send one to, and a panel that pretended otherwise would be showing you an answer we made up. Install it and call it from your own client — the Installation tab has the entry for each one.
What it does
Google, Bing, DuckDuckGo and Google Maps search results, forwarded from the Litescrape API.
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user serpbridge-mcp '--' npx -y @reneturcios/serpbridge mcp
# 2 — ask your agent for something these tools do
# google_search, bing_search, duckduckgo_search and 1 more
Serpbridge costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Collected from a public index. Nobody has claimed this account, so nothing here was written by its author — claim it if it is yours.
Where are you running it?
Every route below installs the same thing and ends at the same approval screen. This one runs on your machine: your client starts Serpbridge as a process under your own user, with your files and your network, so the tool surface below is what it can do to your computer rather than to a server somewhere else. Its scan result is on the Security tab — read the surface before you approve it.
This is a public server: you run it yourself and this marketplace is not in the path. You need Node.js — the current LTS (24) is the safe choice; npx comes with it. Claude Code registers it in one command; --scope user makes it available in every project. The '--' is in quotes so that PowerShell passes it on.
claude mcp add --scope user serpbridge-mcp '--' npx -y @reneturcios/serpbridge mcpReconnect, or start a new session, and the tools appear in the model’s tool list.
One config entry your client uses to start the process locally. A local server runs with your file system and your network, which is why it is priced without metering.
4 tools, with what each one reads, writes and reaches shown before you agree — the same list on every route above. Read the tool surface.
Tool surface
What the model actually sees. Descriptions are diffed on every release — see version history.
Read from its source code. A server that builds its tool list when it starts can list more tools than this.
Get Google search results, including the AI Overview and People also ask.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| page_token | no | Opaque token from pagination.next_page_token. Send it alone to fetch the next page with the original search parameters. Prefer page_token to repeating the same search with start=10, start=20, or anot | |
| q | no | Google Search query, up to 2,048 characters. Required unless ludocid, kgmid, or page_token is supplied. | |
| fast_mode | boolean | no | Return organic results only, skipping AI Overview and all other result groups. Keeps search metadata and parameters. Omits cross-module position_absolute. |
| location | no | Human-readable city-level location encoded to UULE by Litescrape. Conflicts with uule and lat/lon. | |
| uule | no | Pre-encoded Google location value. Conflicts with location and lat/lon. | |
| lat | no | Latitude from -90 through 90. Requires lon and conflicts with location and uule. | |
| lon | no | Longitude from -180 through 180. Requires lat and conflicts with location and uule. | |
| radius | no | Decimal search-bias radius in meters. Requires location or lat/lon. Desktop: 1–199; tablet/mobile: 1–1,000. | |
| ludocid | no | Google local CID used to search for one entity. | |
| lsig | no | Opaque Knowledge Graph or local-pack signature. | |
| kgmid | no | Google Knowledge Graph machine ID, such as /m/0k8z. | |
| si | no | Opaque cached Google search context. | |
| ibp | no | Google layout or expansion control. | |
| uds | no | Opaque Google filter token. | |
| color_scheme | no | Google result color presentation. | |
| google_domain | string | no | Results use google.com. Use gl, hl, cr, lr or geography for localization. |
| gl | no | Lowercase two-letter country localization code. | |
| hl | string | no | Google interface and result language code. |
| cr | no | One or more countryXX restrictions joined with |. | |
| lr | no | One or more lang_xx restrictions joined with |. |
Get Bing web search results.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| q | string | yes | Bing search query after trimming. Accepted length: 1 through 2,048 characters. Control characters are rejected. Bing query syntax such as quoted phrases and exclusions remains available. |
| engine | string | no | Selects the Bing web-search engine. |
| location | no | Named city-level origin used to localize the Bing request. Maximum length: 256 characters. Control characters are rejected. | |
| lat | no | Latitude from -90 through 90. May be supplied independently or with lon. | |
| lon | no | Longitude from -180 through 180. May be supplied independently or with lat. | |
| mkt | no | Case-insensitive Bing language-country market token. Format: two-letter language and country, such as en-US. Cannot be combined with cc. | |
| cc | no | Case-insensitive two-letter country of origin. Format: US or GB. Cannot be combined with mkt. | |
| first | integer | no | One-based native offset for organic results. Range: 1 through 9,007,199,254,740,991. Prefer litescrape_pagination.next when it is returned. |
| safeSearch | string | no | Bing adult-content policy; values are case-insensitive. |
| filters | no | Native Bing display or date filter expression. Maximum length: 8,192 characters. Control characters are rejected. | |
| device | string | no | Selects desktop, tablet, or mobile result behavior. |
| result_groups | array | no | Return only these top-level result groups, e.g. ["organic_results", "knowledge_graph"]; omit for the complete response. search_metadata is always kept |
Get DuckDuckGo web search results.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| q | string | yes | Required search text, up to 500 characters. |
| kl | no | DuckDuckGo region and language token, such as us-en. | |
| safe | string | no | DuckDuckGo safe-search level. |
| df | no | Date window: d, w, m, y, or YYYY-MM-DD..YYYY-MM-DD. | |
| search_assist | boolean | no | Enable DuckDuckGo query assistance; mutually exclusive with m. |
| start | integer | no | Result offset from 0 through 10,000. |
| m | integer | no | Requested result count from 1 through 50; mutually exclusive with search_assist. |
| result_groups | array | no | Return only these top-level result groups, e.g. ["organic_results", "knowledge_graph"]; omit for the complete response. search_metadata is always kept |
Search Google Maps places, or get one place by ID.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| q | no | The business, category, address, or natural-language query to search. Required when type is search. Examples: coffee shops in Austin, dentist near Denver | |
| ll | no | A complete viewport containing latitude, longitude, and either zoom or radius. Format: @lat,lon,14z or @lat,lon,5000m | |
| location | no | A named geographic location resolved by the API. Cannot be combined with ll or lat/lon. Requires z or m. | |
| lat | no | Latitude for a coordinate-based viewport. Must be supplied with lon and either z or m. | |
| lon | no | Longitude for a coordinate-based viewport. Must be supplied with lat and either z or m. | |
| z | no | Google Maps zoom level for location or lat/lon geography. Accepted range: 3 through 30. Cannot be combined with m. | |
| m | no | Search radius in meters for location or lat/lon geography. Accepted range: 1 through 15,028,132. Cannot be combined with z. | |
| nearby | no | Uses the supplied geography as a nearby-search scope. Requires ll, location, or a lat/lon pair. | |
| data | no | A Google Maps protobuf parameter sequence for search filters or an exact place. Maximum length: 8,192 characters. Exact-place data requires type=place. | |
| place_id | no | A Google place identifier used to resolve one exact place or request live foot traffic. Cannot be combined with data_cid or exact-place data on the primary Maps endpoint. It is the only required para | |
| data_cid | no | A decimal Google CID used to resolve one exact place. Cannot be combined with place_id or exact-place data. | |
| type | no | Chooses a result search or an exact place response. Required for query and filter searches. Use place when sending an exact-place data sequence. | |
| hl | no | Language used for Google Maps labels and returned text. Examples: en, es, fr, de | |
| gl | no | Two-letter country code used for regional result localization. Examples: us, gb, ca, au | |
| google_domain | no | Validated Google domain used for the request. Examples: google.com, google.co.uk, google.ca | |
| start | no | Native Maps result offset: 0, 20, 40, and so on. Each request returns up to 20 places and each successful request costs one credit. Use pagination.next when it is present instead of constructing the | |
| min_price | no | Minimum price level for returned places. Must be non-negative and cannot exceed max_price. | |
| max_price | no | Maximum price level for returned places. Must be non-negative and cannot be lower than min_price. | |
| min_rating | no | Preferred minimum Google rating. Google treats this as a relevance preference. | |
| open_state | no | Filters by the current open state. Cannot be combined with open_on_day or open_at_hour. |
- Every tool, no call limit
- No card, and mcprush charges nothing for it
- Source published under MIT
- Runs on your machine — nothing of it reaches our gateway
- Nothing to cap, because mcprush bills nothing
What counts against your monthly calls
| Tool | Unit | Calls used | Out of the allowance |
|---|
Nothing here is billable: mcprush charges nothing to install Serpbridge or to call it, at any volume.
Two independent axes, because powerful and malicious are different questions. The grade is threat only. The capability level is blast radius, and it is never a penalty on the grade — it is priced as one subtract-only term in the score, where you can see it.
| Term | Level | What it prices | Points |
|---|---|---|---|
| capability-exposure | minimal | capability blast radius (minimal) — client exposure if the model is manipulated | −0 |
| verification-discount | repo | publisher verification (public source) — no provenance, but the source is public and inspectable | −1 |
| coverage-honesty | source | inspection depth (source) — how much of the target the scan could see | −0 |
What the scan could actually read
A grade is only as meaningful as its coverage, so the scanner publishes its own depth before it publishes its result.
No tools were enumerated, so prompt-injection, capability and toxic-flow analysis had no tool surface to inspect. To grade a package’s real runtime tools, scan the running server: --command "npx -y <package>".
Capability — what it could do if the model were manipulated
Tags derived from each tool’s schema and the implementation, not from what the tool calls itself. minimal is the level these add up to.
| Tool | Capability tags | Why the tag was assigned |
|---|---|---|
| No per-tool rows for this release — see the note below. | ||
Toxic-flow graph
The lethal trifecta, checked as a graph rather than as a checklist: untrusted input, a sensitive source and an external sink have to meet before there is a path worth worrying about.
The public result for this release does not print the flow graph, so there is nothing to show here. That is not the same as "no paths were found": what the scan did read is above, under coverage.
Supply chain and provenance
This is the first scan of this surface here, so there is nothing yet to compare it against.
Every result on this tab comes from one deterministic pass over the published package — offline, rule by rule, and auditable line by line above. Same methodology version, same bytes, same score.
Release history
Always latest — the install command names no version, so it installs the newest release when it runs; the current release on file is 0.1.0. Pin a release below and the command asks for exactly that one.
No release note was published with this version.
One review per account, from a signed-in account that does not publish this listing. Nothing else is asked — you can say what you think before you install it. Publishers can reply once per review, and a review can be edited or deleted by whoever wrote it.
Any signed-in account that does not publish Serpbridge can review it, once — before installing it or after. A review from an account that has installed it is marked as one.
Nobody has reviewed this listing. The rating on the card is the mean of the reviews written here and nothing else, so there is no rating until somebody writes the first — which takes a signed-in account that does not publish it, and nothing else.