MoongraphMoongraph

Agent

Ask questions about your documents and get cited answers.

Agent

The Agent is Moongraph's AI chat interface. Ask questions in natural language and get answers grounded in your documents with citations linking back to sources.

Getting started

How it works

When you ask a question:

  1. Query analysis — The Agent decides which tools to use
  2. Tool execution — Retrieval and analysis tools run (shown as pills in the UI)
  3. Context assembly — Relevant content is gathered from your documents
  4. Answer generation — The LLM generates a response using document context
  5. Citation linking — Claims are connected to source chunks

This is Retrieval-Augmented Generation (RAG). The AI answers based on your actual documents, not general knowledge.

Available tools

The Agent has 12 specialized tools:

CategoryTools
Document retrievalretrieve_chunks, retrieve_document, list_documents, search_documents
Document analysisget_document_stats, get_document_summary
Knowledge graphslist_graphs, graph_overview, entity_search, get_entity_sources, knowledge_graph_query
Visualizationget_visualization_help

See Agent Tools Reference for complete parameter documentation.

Citations

Every factual claim includes citation markers. Click a citation to:

  • See the exact source chunk
  • Open the source document
  • Navigate to the relevant page

Visualizations

Ask for diagrams and the Agent generates interactive visualizations:

  • Mermaid diagrams — Flowcharts, timelines, sequence diagrams, state diagrams
  • Data charts — Bar, line, area, pie, scatter plots
  • Mind maps — Hierarchical concept breakdowns

Visualizations render directly in the chat. See Generate Visualizations.

Scope and access

The Agent queries documents based on your current scope:

  • Personal scope — Your personal documents only
  • Team scope — Team documents and shared content

Switch scopes in the sidebar to change what's accessible.

Limitations

  • No real-time data — The Agent knows what's in your uploaded documents
  • Context limits — Very long documents may be truncated
  • Extraction quality — Answers depend on underlying document quality
  • Visualization data — Charts require extractable data from documents

Learn more

On this page