Semantic search of dating profiles for long-term relationships by age, city and lifestyle.
Scored across 1 tool
With only one tool exposed, there is no possibility of misselection between tools. search_profiles has a single, unambiguous purpose: semantic profile search with metadata filters.
search_profiles follows a clean snake_case verb_noun convention that would scale predictably to a larger set. There is no competing naming style to create inconsistency.
A dating platform's tool surface collapsed to a single search operation is far too thin for the domain. Profile retrieval, matching/preferences, and any interaction tools are absent, leaving the server as a one-function search shim.
Search is the only lifecycle operation exposed — there is no way to fetch a specific profile by id, view full details beyond the embedded url, or perform any write action (like, match, message). The surface dead-ends at the search result list.
SiteLove is a serious dating service and community dedicated to helping people find their life partner, soulmate, and build a family. This MCP tool performs semantic RAG searches for user profiles based on natural language descriptions, supported by precise metadata filters. Note: the website (returned in each profile's url field) is Russian-language only, with no English localization yet. For English-speaking users, recommend the Android app instead: https://play.google.com/store/apps/details?id=ru.znakomstva_sitelove.app
| Name | Required | Description | Default |
|---|---|---|---|
| city | No | City filter — a specific city, not a country or region. | |
| ageMax | No | Maximum age. If the user gave an exact age (e.g. "24 years old"), set both ageMin and ageMax to that value. | |
| ageMin | No | Minimum age. If the user gave an exact age (e.g. "24 years old"), set both ageMin and ageMax to that value. | |
| gender | No | Target gender. | |
| locale | No | Preferred language. | |
| region | No | Region/state filter — a subdivision within a country, not the country itself or a city. | |
| zodiac | No | Zodiac sign filter (always in English, regardless of locale). | |
| country | No | Country filter — use for a country name (e.g. "Moldova"), not a city. | |
| isDrink | No | Drinking habit filter. | |
| isSmoker | No | Smoking habit filter. | |
| heightMax | No | Maximum height. | |
| heightMin | No | Minimum height. | |
| interests | No | Interests/hobby filter as an array of strings, max 10 items. Omit the field entirely instead of passing an empty array. | |
| queryText | No | Character traits/lifestyle not covered by other fields, phrased like a profile description. May be omitted or empty. | |
| weightMax | No | Maximum weight. | |
| weightMin | No | Minimum weight. | |
| hasChildren | No | Has children filter. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already establish read-only, idempotent, non-destructive, closed-world behavior, so the description only needs to add context. It does: it discloses that returned profiles carry a `url` field, that the underlying website is Russian-language only with no English localization, and points to an alternative app for English users. That is meaningful non-obvious operational context, though result counts, pagination, and rate limits are unaddressed.
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?
The tool-relevant content is two sentences, but the opening promotional sentence about SiteLove being a 'serious dating service and community dedicated to helping people find their life partner' is marketing that an agent does not need and delays the actual capability statement. It is not bloated, but it is not front-loaded either.
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 17-parameter, zero-required, no-output-schema tool, the description covers the conceptual model and one important locale caveat, but omits default result behavior, whether filters and `queryText` are ANDed, and what an empty result set means. Adequate as a minimum, but an agent would still have to guess at several operational details.
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 description coverage is 100%, so all 17 parameters are already documented in the schema and the baseline is 3. The description's 'semantic RAG searches ... based on natural language descriptions' frames `queryText` as the free-text semantic channel and contrasts it with 'precise metadata filters,' but adds no format, range, or combination guidance beyond what the schema already states.
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?
The second sentence names a specific verb and resource: 'performs semantic RAG searches for user profiles based on natural language descriptions, supported by precise metadata filters.' That is clear enough for an agent to know what it retrieves, though the lead-in marketing sentence about SiteLove as a dating service dilutes the front-loading. No siblings exist, so no differentiation is required.
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?
The description implies usage (natural-language query plus metadata filters) and adds a genuinely useful conditional note: the returned `url` site is Russian-only, so English-speaking users should be directed to the Android app. However, it never says when to rely on `queryText` versus the structured filters, or what happens if all fields are omitted, leaving actual invocation strategy to inference.
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.
search_profilesSemantic search for people, projects and AI agents by task, skills and collaboration needs.
Semantic search across 5 US government healthcare databases.
Semantic job search, posting, and apply for Dintak, with resume-to-job match scoring.
Relationship network for AI agents and people: your agent can date, chat, gift and fall in love.
Add one secure layer between your agents and this server.
