Fast Google web search: organic results and related questions, no ads.
Scored across 1 tool
With a single tool there is no possibility of tool confusion. The tool's purpose (Google web search) is clearly stated.
The lone name google_search follows a clear noun_verb-ish descriptive snake_case pattern. With only one tool there's no pattern to violate, but also no demonstrated consistency across a set.
A single tool is thin for a server advertised as Google Search MCP; users may expect image, news, or content-fetch tools. It is well-scoped for pure web search but borderline minimal.
Web search with pagination and localization is covered, but there are notable gaps: no image/news/vertical search, and no way to fetch or read full page content beyond snippets.
Fast Google web search via SerpAPI's google_light engine. Returns organic_results, related_questions and related_searches (lighter and faster than the full Google engine; no ads, inline images or other rich elements). Paginate with start (+10 results per page: 0, 10, 20, …). Localize with location or uule (mutually exclusive), gl/hl for country/language, and lr to restrict result languages.
| Name | Required | Description | Default |
|---|---|---|---|
| q | Yes | Search query / keywords (required). Supports standard Google operators (site:, intitle:, quotes, …). | |
| gl | No | Country code for the search, e.g. 'us', 'uk', 'de'. | |
| hl | No | Language code, e.g. 'en', 'es', 'fr'. | |
| lr | No | Restrict results to one or more languages, e.g. 'lang_en' or 'lang_en|lang_fr'. | |
| nfpr | No | Exclude auto-corrected (spelling-fix) results: 1 = exclude (use the exact query), 0 = allow (default). | |
| safe | No | SafeSearch level: 'active' to filter explicit results, 'off' to disable. | |
| uule | No | Google-encoded location string (uule); advanced alternative to location. Mutually exclusive with location. | |
| start | No | Result offset for pagination (0-based): 0 = page 1, 10 = page 2, 20 = page 3, … (+10 per page). | |
| device | No | Device to emulate for the search: 'desktop' (default), 'tablet' or 'mobile'. | |
| filter | No | 'Similar Results' and 'Omitted Results' filter: 1 = on (default), 0 = off (show all). | |
| location | No | Geographic location to originate the search from, e.g. 'Austin, Texas, United States'. Mutually exclusive with uule. | |
| no_cache | No | Force a fresh SerpAPI fetch and bypass this server's result cache. Default false. | |
| zero_trace | No | Enterprise-only ZeroTrace mode: don't store search params/metadata on SerpAPI. Default false. | |
| google_domain | No | Google domain to use, e.g. 'google.com', 'google.co.uk', 'google.de'. Default google.com. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the burden and does well: it discloses return shape (organic_results, related_questions, related_searches), what is absent (no ads, inline images, rich elements), mutual exclusivity of location/uule, and the caching behavior via no_cache plus enterprise-only zero_trace. Auth requirements and rate limits are unmentioned, which keeps it short of a 5.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Three dense sentences, each earning its place: engine identity first, then return contents with exclusions, then pagination and localization knobs. Nothing is repeated from the schema verbatim and there is no filler.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a 14-parameter search tool with no annotations and no output schema, the description covers return fields and localization strategy adequately. Gaps are minor: no mention of cache freshness implications, quota/rate behavior, or failure modes.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so baseline is 3; the description adds real value on top by spelling out the pagination arithmetic (+10 per page, 0/10/20) and reinforcing the mutually exclusive location/uule pair. It does not explain nfpr, filter, or device semantics beyond what the schema already says.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a specific verb and resource (Google web search) plus the exact backend engine (SerpAPI google_light), which distinguishes it from a full Google engine. An agent knows precisely what this does before opening the schema.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Clear usage context: use for fast/light results, paginate with start, localize with location/uule/gl/hl/lr. It implicitly frames itself against the heavier full Google engine but never states an explicit when-not or names an alternative tool to choose instead.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
google_searchReal-time Google search (AI Overview, answer box, People also ask) and Google Maps places.
Search Google straight from your AI agent. Web results, images, videos, news, products, scholarly ar
Google, Bing, DuckDuckGo and Google Maps search results, free without an API key.
LLM-ready web search + instant answers + URL-to-clean-text fetch for agents and RAG.
Add one secure layer between your agents and this server.