Introspect RDF datasets and retrieve information via SPARQL with GenAI natural language queries.
npx skills add https://github.com/sparq-org/sparq --skill genai-retrieval
Please help me install this Agent Skill. Skill: genai-retrieval Source: https://github.com/sparq-org/sparq/tree/main/skills/genai-retrieval Command: npx skills add https://github.com/sparq-org/sparq --skill genai-retrieval
This Skill automates the process of understanding and retrieving information from RDF datasets using SPARQL, enhanced with natural language processing capabilities provided by Generative AI.
Load an RDF dataset with 'sparq-core::Graph::load_str("data.ttl", "turtle")?;'. Use the 'sparq_introspect::Introspection::build(&graph)' to generate a schema card. To query, initialize an NLQ with 'sparq_nlq::Nlq::new(&graph, Box::new(ReplayLlm::from_file("tests/fixtures/olympics_replay.json")?));' and then call 'ask("What are the average grades of students?")?;'
High-intent search queries and answers about installing and using this skill.
You can query RDF datasets using natural language by leveraging NLQ integration, which translates natural language questions into SPARQL queries. This allows you to ask questions directly against RDF graphs without writing SPARQL manually.
RDF dataset introspection is the automated process of generating schema cards, VoID representations, and SHACL node shapes for datasets. It helps with data retrieval by providing structured summaries of unknown datasets, enabling token-budgeted retrieval contexts.
Yes, you can retrieve data from unknown RDF datasets without prior schema knowledge. The Skill builds token-budgeted retrieval contexts using NL to SPARQL loops, allowing ad-hoc querying of unfamiliar RDF graphs.
Yes, sparq-core, sparq-introspect, and sparq-nlq are required dependencies. You need sparq-core to load the RDF graph and sparq-nlq to initialize the natural language query interface for SPARQL-based retrieval.
The best way to generate a schema card for a Turtle dataset is to load the data using a graph loader and then run an introspection build process. This automatically generates schema cards, VoID representations, and SHACL node shapes.
Natural language queries may fail if the NLQ interface is not properly initialized with the loaded graph and a language model. You must initialize the NLQ instance with the graph object and an LLM replay file before calling the ask function.
sparq-orgIngest RDF data into sparq Graphs from Turtle, N-Triples, N-Quads, and TriG formats.
zot24Provide persistent memory and context retrieval for LLM applications.
sparq-orgExecute SPARQL 1.1/1.2 queries on RDF data in the sparq engine.
sparq-orgQuery SPARQL RDF datasets via the Model Context Protocol.
