Skip to main content
Glama
gca-global

Qobrix CRM MCP Server

by gca-global

qobrix_search_projects

Search real-estate projects with precise filters for must-haves and weighted preferences for nice-to-haves, returning top matches ranked by relevance.

Instructions

Highly relevant project (development) search for free-language demand (F1-optimized). TWO-TIER RECIPE: (1) search = hard DSL must-haves (server filter → precision). (2) boost[] = soft weighted nice-to-haves scored over up to max_scan candidates (recall + ranking). (3) limit = how many ranked rows to return (default 10, max 100). With boost: top-N with _relevance and _matched; pagination.mode='ranked'. Without boost: fast path single cached page; pagination.mode='fast'. Call qobrix_search_dsl_help({resource:'Projects'}) for DSL + field cheatsheet. Example: search='city contains "Paphos"', boost=[{field:'starting_price_from',op:'<=',value:400000,weight:2},{field:'construction_stage',op:'==',value:'under_construction',weight:1}], limit=10, max_scan=150. When expand=true or media=true, max_scan is auto-capped at 100. If status='result_too_large' with _refine_required, ask the user to narrow then retry. All upstream pages are response-cached.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
pageNoPage number (default 1). Used on the fast path (no boost). Ignored when boost is set (ranking returns a single top-N page).
sortNoSort by field name (maps to Qobrix OpenAPI sort[]). Prefix with - for descending. Comma-separated for multi-key (e.g. '-list_selling_price_amount,-created'). Examples: '-created' (newest first), 'name' (alphabetical), '-list_selling_price_amount' (highest list price first).
boostNoSoft relevance criteria (nice-to-haves). Never filters out rows — only ranks them. When present, the tool scans up to max_scan candidates matching `search`, scores each row as the sum of matched clause weights, and returns the top `limit` with _relevance and _matched. Put must-haves in `search`; put preferences here. Example: [{field:'sea_view',op:'==',value:true,weight:3},{field:'bedrooms',op:'>=',value:3,weight:2}].
limitNoHow many results to return (1-100, default 10). With boost: top-N after ranking. Without boost: page size. Raise when the user wants more options; keep low to avoid context overload.
mediaNoInclude inline media on each row. Default false.
expandNoExpand FK references into nested objects. Default false.
fieldsNoLimit response to specific fields only (partial response). Reduces payload size. Example: ['id','name','status','list_selling_price_amount']. Omit to get all fields.
searchNoHard-filter Qobrix search expression (server-side precision). Operators: == != <> < > <= >=, contains, starts with, ends with, in [...], not in, ranges in a..b, and/or/not. Functions: DISTANCE_FROM, IN_POLYGON, TRANSLATED, MIN/MAX, DAYS_AGO(n), MONTHS_AGO(n), DAYS_FROM_NOW(n). Shortcuts: NOW, TODAY, THIS_WEEK, LAST_MONTH, THIS_YEAR, CURRENT_USER. Strings double-quoted; booleans true/false; association paths e.g. SalespersonUsers.Contacts.country. Example: status == "available" and sale_rent == "for_sale" and list_selling_price_amount <= 500000. For the full grammar + field cheatsheets call qobrix_search_dsl_help. For enum values call qobrix_get_field_options; for all fields call qobrix_get_schema.
max_scanNoCandidate pool size when boost is set (default 100, hard cap 500). When expand=true or media=true the effective scan is capped at 100 (pagination.scan_capped_reason='expand/media') to keep latency and payload size safe. Higher improves recall (less chance of missing a good listing) but costs more API pages. Ignored on the fast path (no boost). Each scanned page is response-cached.
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description fully carries behavioral disclosure. It explains ranking behavior, _relevance/_matched fields, pagination modes, auto-capping of max_scan with expand/media, response caching, and result_too_large error handling — all beyond what the schema provides.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is dense but every sentence earns its place, covering purpose, recipe, example, edge cases, and helper references. It is front-loaded with the purpose and uses a clear two-tier recipe structure, making it compact for the complexity it handles.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given 9 parameters, no output schema, and no annotations, this description provides a remarkably complete picture: input semantics, ranking vs fast modes, scan caps, caching, page behavior, and error-refinement guidance. An agent can correctly select and invoke the tool without needing external documentation.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Although schema coverage is 100%, the description adds substantial meaning: it details the search/boost/limit/max_scan interaction, gives concrete examples, clarifies that boost never filters but only ranks, and notes the fast-path behavior with or without boost. This is far beyond the baseline schema descriptions.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description opens with a specific, non-tautological statement: 'Highly relevant project (development) search for free-language demand (F1-optimized).' It clearly identifies the verb (search), resource (projects/developments), and scope (free-language demand), and the distinctive two-tier search/boost recipe further differentiates it from sibling search tools.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It gives clear guidance on how to use the tool internally: 'search = hard DSL must-haves' vs 'boost[] = soft weighted nice-to-haves', and explains fast vs ranked paths. However, it does not explicitly tell when to prefer this tool over alternatives like qobrix_list_projects or qobrix_get_project, so it stops short of a 5.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/gca-global/qobrix-crm-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server