VeritasGraph
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
VeritasGraph answers questions from a local knowledge graph with verifiable citations. Before answering a factual question, call veritasgraph_check_answerable; if it is not answerable, say so instead of guessing. Use veritasgraph_query and quote its [doc#chunk] citations; treat answers with grounded=false or abstained=true as unverified. Use mode='governed' when the user needs answers restricted to certified business terms. Glossary terms and types are in veritasgraph_get_ontology.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
| resources | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| veritasgraph_check_answerableA | Check whether the knowledge graph covers a question before asking it: returns matched entities and glossary terms, coverage and a reason. No LLM call. |
| veritasgraph_queryB | Ask a question and get a graph-grounded, multi-hop answer with verifiable [doc#chunk] citations, a grounded flag, the reasoning path and an answer_id receipt. Abstains when the graph has no evidence. |
| veritasgraph_search_entitiesA | Retrieve the subgraph most relevant to a query (entities, relationships, seeds) without invoking the LLM. Fast graph lookup. |
| veritasgraph_get_entityB | One entity (by id or name) with neighbours, source chunks and glossary terms. |
| veritasgraph_get_graphA | Return the full knowledge graph: all nodes, edges, and stats. |
| veritasgraph_get_ontologyB | The governed vocabulary: entity types, predicates (with domain/range) and glossary terms, each with status proposed / certified / deprecated. |
| veritasgraph_search_termsB | Search the business glossary by name, synonym or definition. |
| veritasgraph_get_termB | One glossary term (by id or name) with its definition, synonyms, status and linked entities. |
| veritasgraph_get_answer_receiptA | Fetch the receipt for a previous answer_id: evidence ids, grounding, the graph version it used, whether the graph changed since, and feedback. |
| veritasgraph_compile_statusB | Last ontology compile run and the number of open curation suggestions. |
| veritasgraph_structured_searchB | Search the structured-data catalog (terms, measures, tables, columns) by meaning and name. No SQL is run. |
| veritasgraph_structured_get_termA | One structured-data business term by id or name, with definition, synonyms and status. |
| veritasgraph_structured_list_measuresC | Named SQL measures with their expression, table and review status. |
| veritasgraph_structured_askA | Answer a question from connected databases with one validated, read-only, row-capped SQL query. Returns the SQL, rows and a receipt; refuses with status no_coverage when the catalog does not cover the question. |
| veritasgraph_ingest_documentA | Ingest a document into the VeritasGraph knowledge graph. Chunks the text, extracts entities and relationships with a local model, and records the source chunk behind every node/edge for verifiable attribution. |
| veritasgraph_import_graphA | Import a pre-built knowledge graph produced by another tool (Graphify graph.json, Understand-Anything KnowledgeGraph, AI Atlas taxonomy JSON, or any node/edge JSON). Imported nodes/edges are tagged with source_type provenance (curated/extracted/inferred) and their original id/version so curated content stays distinguishable. |
| veritasgraph_propose_termB | Propose a business glossary term (status 'proposed' until an admin certifies it), optionally linked to entity ids. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| VeritasGraph knowledge graph | Live snapshot of the local knowledge graph (nodes, edges, stats). |
| VeritasGraph statistics | Counts of entities, relationships, and source chunks. |
| VeritasGraph ontology | Entity types, predicates and glossary terms with their status. |
TDQS
Scored across 17 tools
Most tools have clearly distinct targets: graph QA (query) versus SQL QA (structured_ask), full graph (get_graph) versus subgraph (search_entities) versus single entity (get_entity), each at a distinct granularity. The weakest boundary is glossary terms (search_terms/get_term) versus structured-data catalog terms (structured_search/structured_get_term), which share vocabulary and could be misselected by an agent despite the 'structured_' prefix hinting at the distinction.
The set is overwhelmingly verb_noun (get_entity, search_terms, propose_term, ingest_document, import_graph), with a coherent 'structured_' namespace for the SQL sub-domain. Minor deviations: bare 'query' lacks a noun, 'compile_status' is a noun phrase with no verb, and the structured_* tools put the namespace before the verb, inverting the dominant pattern.
17 tools spans three coherent sub-surfaces (graph QA/verification, ontology+glossary curation, structured-data access) plus ingestion, and each tool maps to a real operation. It is on the heavy side of the ideal band and a few read tools (e.g. get_term vs search_terms) could be merged, but nothing feels gratuitous.
Core workflows are covered end-to-end: ingest, query with citations, verify via receipts and check_answerable, inspect ontology and entities, and query structured data. However, the curation lifecycle is a dead end — propose_term exists and compile_status reports open curation suggestions, but there is no tool to list, certify, update, or delete terms/suggestions, and receipts expose feedback with no way to submit it.