LHO MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| LHO_PORT | No | Port of the chat page. | 8000 |
| LHO_MODEL | No | Any Ollama model that supports tool calling. | qwen2.5:7b |
| LHO_ENABLE_SPARQL | No | Set to 1 to add the run_sparql tool. | off |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
Tools for synthetic diagnostic test records of cattle (six laboratories) and pigs (two laboratories), described with the Livestock Health Ontology (LHO). Use test_statistics for any number. Use search_terms or describe_term for the meaning of a term. Never invent numbers.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| graph_overviewB | What the graph contains: number of records, date range, and the valid values of every dimension (species, pathogen, country, lab, breed, sample_type, diagnostic_test, production_stage). |
| test_statisticsA | Count test records: tested, positive, negative and positivity_percent. Use this for every question about numbers, percentages, comparisons or trends. Leave a filter empty when the question does not mention it. group_by: leave EMPTY for one number or one rate ("how many", "what is the positivity"). Set it ONLY to compare groups ("which ... highest, lowest, most"): one of species, pathogen, country, lab, breed, sample_type, diagnostic_test, production_stage, year, year_month, month_of_year. The result then has "overall" (the whole selection), one row per group, and highest_positivity, lowest_positivity, most_tested, most_positive. "Which pathogen has the highest positivity?" -> group_by="pathogen". "Which laboratory tested the most samples?" -> group_by="lab". "Which breed ... for Mycoplasma bovis?" -> group_by="breed", pathogen="Mycoplasma bovis" (do not also fill the breed filter). pathogen: cattle: BRSV, BCV, PI3, "Pasteurella multocida", "Mannheimia haemolytica", "Histophilus somni", "Mycoplasma bovis". Pigs: PRRS, "Swine influenza virus", "Mycoplasma hyopneumoniae", "Lawsonia intracellularis". Or a group: "bacteria", "virus". country: Belgium, France, Ireland, The Netherlands (cattle), Spain (pigs). lab: 1 to 6 (cattle), 7 and 8 (pigs). breed: Beef, Dairy, Suckler, Veal, Mixed, Unknown (cattle), Meat (pigs). sample_type: Autopsy, BAL, Swab, TTA, Tissue, Feces, Fluid, Unknown, Missing. diagnostic_test: PCR, Culture, NPS. production_stage (pigs only): Suckling, Nursery, Weaning, Finishing, Breeding. year: four digits, for example 2024. month: "2022-01" for one month of one year, or 1 to 12 for that month in every year. date_from, date_to: 2024, 2024-03 or 2024-03-15 (both ends included). species: leave EMPTY unless the question itself names an animal. Empty means both species, which is what "overall", "in total" and "all tests" ask for. Fill "Pig" only for a question about pigs, "Cattle" only for a question about cattle. Several values for one filter may be given with commas: country="Belgium, France". Synonyms and other languages work (for example "kweek" for Culture). |
| list_valuesA | List the valid values of one dimension with the number of tests for each. dimension: species, pathogen, country, lab, breed, sample_type, diagnostic_test or production_stage. |
| search_termsA | Find ontology terms by name, abbreviation or synonym, in any language. Use it when you are not sure which term the user means. |
| describe_termA | Definition of one ontology term: what it is, its parent classes, synonyms, names in other languages and its relations (which disease a pathogen causes, which species it affects, the neighbours of a country). Pass ONE term, not a sentence. Use it for "What is BRSV?", "Is PI3 a virus?", "Which countries border Belgium?" (term: Belgium). |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 5 tools
Each tool has a fairly distinct role: graph_overview for corpus metadata, test_statistics for numeric answers, list_values for enumerating a dimension, search_terms vs describe_term for ontology lookup vs detail. The main overlap is between graph_overview (which lists valid values of every dimension) and list_values (which lists values of one dimension with counts), so an agent may hesitate between them. Other boundaries are clear.
All names are lowercase snake_case with two words, which is readable and predictable. However, the convention is mixed: list_values/search_terms/describe_term use verb_noun while graph_overview/test_statistics use noun-first forms. Minor deviation, not confusing.
Five tools for a focused domain (aggregate statistics plus ontology lookup over a diagnostic-test graph) is well scoped. Each tool covers a distinct capability: metadata, querying, value discovery, ontology search, ontology description. Nothing feels redundant or missing at the count level.
The surface covers aggregate querying, dimension discovery, and ontology term lookup, but there is no way to combine two dimensions in one query (e.g., pathogen by country), no trend/time-series tool beyond group_by year, and no record-level or free-form query escape hatch. Agents hitting a cross-tab or raw-record question will hit a dead end.